In January 2025, Workday announced it was rolling TechWolf out across its own 20,400 employees. Its chief people officer, Ashley Goldsmith, said the partnership would turbocharge Workday’s existing skills offering.

Head over heels, basically.

On Tuesday, SAP agreed to buy TechWolf. Workday put money into TechWolf’s Series B in 2024. So did SAP and ServiceNow, and only SAP bought the company.

So Workday’s skills-inference partner now belongs to SAP. Mad world.

Josh Bersin wrote this spring that the world is moving from “systems of record” to “systems of context.” I’m borrowing the line because it describes this deal better than anything I could come up with.

For a long time the fight was over where your data lives. Now it’s over what an agent asks when it needs to know who does what.

SAP’s own pitch for TechWolf is about agents. Manoj Swaminathan, SAP’s president and chief product officer for the Autonomous Suite, said the context graph would ground agent queries and make token use more efficient. He also pointed to cheaper deployment of workforce agents. Skills-based hiring comes later in the list.

Bersin’s version of context is the business rules and objects your HCM already holds, which is why he points to Workday’s Business Process Framework as an existing semantic layer. Foundation Capital popularized “context graph” in a paper on decision traces, meaning a record of what got decided and why. Its pitch is that AI agent startups building those graphs gain a structural advantage. TechWolf’s version is evidence of what work people actually do and which skills it takes. SAP bought that third kind.

I’d argue the big vendors have split into two camps on where context comes from. Workday is building it on top of its own transaction data. Gerrit Kazmaier told Bersin that Workday’s biggest accuracy gains came from knowledge graphs and context engineering, not bigger models. What Workday has bought is the front door and the plumbing, including Sana and Pipedream (and Paradox, for recruiting).

SAP is buying the context layers themselves. It bought Reltio in March. Dremio and Prior Labs followed in May, and now TechWolf. Constellation’s read on the Dremio and Prior Labs deals was that SAP may be admitting it’s behind.

That’s the skeptical read, but it’s fair.

ServiceNow sits between the camps. It launched a Context Engine in April, assembled partly from companies it bought, and it’s a TechWolf investor and customer too. Microsoft built its own, called Work IQ, and opened up the APIs in June.

Gloat launched Loomra in March, a context layer that sits on top of whichever HCM you run, and Bersin said he wasn’t aware of any HCM vendor that had built one yet. Celonis, the process mining company, bought Ikigai Labs and launched its own Context Model.

Both camps have an agent registry now, so that’s not where they differ. Workday’s Agent System of Record is live, and SAP’s AI Agent Hub is rolling out. Where the context comes from is the difference.

If you’re renewing an HCM contract, ask which camp your vendor is in. Ask where the skills and work data behind its agents comes from. If the answer is a partner, ask what your contract says if that partner gets bought.

Thomas Otter, who invested in TechWolf through Acadian, wrote in May that he didn’t expect the transaction-layer incumbents to win the work context graph. In his view the layer has to bridge ecosystems and their traditional antagonists, and he said SAP was mired in API tollbooth confusion. The day SAP announced the deal, he called it a vindication of Acadian’s thesis.

He may be right both times, but it is a tough fence to walk. A layer that has to work across rival HCM vendors gets harder to run once one of them owns it.

I wrote about the deal itself, and why I worry about TechWolf staying independent, over on the 3Sixty Insights blog. Take a read if this sort of stuff matters to you.

What else is going on this week?

• The September 2026 Jobs Report: Here’s the Breakdown. Keirsten Greggs has a great rundown of the jobs numbers: 29,000 jobs added, unemployment up to 4.2%, and Black unemployment up a full point to 7.0%.

• Why Writing a Thriller Made Me a Sharper CEO. I love this from Donald Thompson, who says writing a political thriller sharpened how he handles the rest of his work.

• Punk Rock HR: Job Searching With Stephanie Fackrell. Stefanie Fackrell tells Laurie Ruettimann‘s Punk Rock HR audience to spend about two hours a day on job boards instead of eight. Good advice here.

• Trust Stories: The Orks and the Ring. Skepticism is warranted. Kristy McCann says the AI accord seven men signed at the White House last week has no enforcement mechanism.

• “Human in the Loop” Is Starting to Sound Suspiciously Like “Human on the Hook”. Francesca Ranieri cites IBM’s finding that 43% of employees say they take the blame when AI goes wrong. Still, don’t put your name to work you can’t back up.

• LinkedIn’s Hiring Assistant 2 Aims to “Just Be Smarter”. David Manaster says Hiring Assistant 2, due in November, will fill in a sparse search from the recruiter’s and company’s hiring history so it does what you meant. If only that worked for people.

• 4 HCM Industry Trends: Service, Compliance, Implementation, and Why 49% of Buyers Would Pay More for Better Support. Good stuff from my colleague Dylan Teggart: 49% of 524 U.S. business leaders in the 3Sixty Insights survey with Insperity would pay more for better human-led HCM support.

• Complaining Is Not A Talent Strategy. Brian Fink says TA is great at diagnosing why creative strategy gets undermined and bad at doing anything about it. A way to rethink how TA makes the case.

• Stop Filling Jobs. Start Building Capability. Kevin Wheeler opens with work no longer happening in neat, permanent boxes, then tells recruiters a permanent hire is only one of four options. The boxes were never permanent, though.

• What Is an AI Agent? And Why Aren’t They All Created Equal?. A good primer for those getting into the agentic game: Jason Averbook walks through four kinds of agents using a Chicago business trip.

• The Future Has a Demolition Phase. Jess Von Bank says this is a rip and retire moment, not a rip and replace one: stop rushing to protect jobs and rush to protect people. Preserve people, not roles, and I love it.

• Have Your Agent Call My Agent. Matt Alder imagines a candidate’s AI agent and an employer’s agent handling the front end of hiring between them. I’m not sure we’re ready for this.

• Two Days in Amsterdam: Notes from Intelligence Group’s Talent Strategy & Intelligence Event. The Talent Intelligence Collective’s recap of Intelligence Group’s Amsterdam event has Lisa K. Simon of Revelio Labs saying employers now need roughly six job postings to make one hire. Always good stuff.

• Surveillance Pricing Is One Way AI Is Destroying Us. Come for the golf, stay for the surveillance. Mike Wood goes from picking up driving range balls to a Big Mac in Fresno at $5.69 in one store and $6.89 two miles away.

• Open Enrollment Is a Foreign Language. Robin Schooling traces why HR ended up translating open enrollment to a World War II wage freeze, and wonders whether HR’s fluency is part of the solution or one of the reasons we’re still stuck. Good questions here.

• 15 Workforce and HR Leaders on What Executives Are Getting Wrong About AI and the Future of Work. Dan Schawbel‘s roundup has great insights, including Dean Carter of Instill warning that agents with no shared norms risk an organizational Lord of the Flies.

• Leadership Through Human Connection. Mervyn Dinnen gives a nice rundown of O.C. Tanner’s Influence Greatness event, including Starbucks reporting a 375% increase in recognition in the first month after adding recognition to its My Daily tool.

• A New Approach to Goal-Setting: We Are *Way* Overdue. I like this take from Katrina Kibben, who sets annual goals as more or less of something rather than a number.

• Apparently, Iced Coffee Can Cost You the Job Now. This story is so dumb. Shannon Pritchett answers the claim that iced coffee in an interview looks like running errands with the obvious: an interview is an errand.

• I Tested Local LLMs for HR. Here’s What I Learned.. I’ve been curious about this too, and Anita Lettink found a Gemma model beat the rest, proprietary ones included, on her Dutch 30% tax ruling question.

• Get in the Right Room. Seth McColley argues a $50M CEO may have more in common with another $50M CEO in a different industry than with a $2M or $2B one.

• What’s on Your Unofficial CV?. Good advice from Damon Klotz, who asks team leads to find out what’s on each person’s unofficial CV, the skills and stories their role doesn’t see yet.

• Thank You, Anne Fulton. Happy trails to Anne Fulton, one of the best, who is handing the CEO role at Fuel50 to co-founder Jo Mills.

• I’m Excited to Announce the Launch of HRME2. Congrats to Howard Tarnoff on launching HRME2, a customer success advisory practice for HCM companies.

In the news

• Unit4 to Acquire Napta and Accelerate Skills-Based Resource Management for Professional & Business Services Firms via Unit4

• How Pearson Will Advance AI Skills With Workera Acquisition via Technology Magazine

• Metaview Raises $60m to Put AI Agents Across the Whole Hiring Process via The Next Web

• Trump H-1B Fee Proposal Tees Up Fight Over Congress’ Authority via Bloomberg Law

• Newsom Signs Law Barring Sole Use of AI in Hiring, Firing Decisions via UPI

• EEOC Sues Harvard for Records in I&D Discrimination Probe via SHRM

• Is EEO-1 Done? What the EEOC’s Proposed Rule Means for Employers via HR Defense

• Booking Goes to Court to Force Staff to Go to the Office via DutchNews.nl

• Why the US Economy Looks Strong but Feels Bad to Many: Explainer via Bloomberg Law

• Workday Global Workforce Report: AI Is Rewriting Jobs More Than It’s Cutting Them via Stock Titan

• Apple’s Reported Overhaul Highlights the Work Behind a Leaner Structure via HR Executive

• AI Is Fueling Union Interest, but Only 9% of Orgs Are Ready via HR Executive

• 6 in 10 Employers Fall Short on Women’s Progression or Retention via HR Grapevine

• Survey: CHRO Confidence Remains in Positive Territory, but Continues to Inch Down via The Conference Board

• Forget ‘Coffee Badging’: ‘Calendar Stalkers’ Game Return to Office via Money Talks News

• With the Iced Coffee Debate, Gen Z Cultural Differences Come to a Head via HR Dive

• PwC Survey on Workforce, AI Skills & More via Bernews

• AI Skills Are Boosting IT Pay, but Mostly for Senior Roles via CIO

• ‘It Was the AI, Not Me’ via Canadian HR Reporter

Have a great rest of your week!

I try not to be too self-referential here but it seems silly not to write that I’ve joined up with 3Sixty Insights as a principal analyst. I should also tell you why and what that means.

I’ve got what some might call an untraditional background. In short, I’ve done everything possible to not create a sensible career path, including going out on my own. I like that.

But that’s also why going down a traditional analyst path isn’t easy. I don’t have the educational pedigree, the proof in public, or the personality to take on that sort of role. The work I do is serious and I care about it deeply, but I don’t take myself that seriously. I am opinionated, but I also know stories and the Truth™ always have different sides and angles to explore. I do not think I have all the answers.

So, yeah, I wasn’t getting calls from Gartner.

I also really like what I am doing now. I like working on my own firm with my own ideas. I like working with the folks at Rep Cap. I like writing for publications like Reworked. I like writing here.

What I really liked is independence. And that’s what attracted me to 3Sixty Insights.

But other than embracing my weird background and my desire to continue what I am doing (and more), there are three other pieces that came together to make it work.

One is that they want me to continue to be opinionated about the market and continue growing my voice in a very crowded industry. That helps both of us. I really didn’t want to yield my voice behind a brand, and at least in this sort of intellectual work, people really care about trusting the voices they are reading.

Two is that I am really aligned with 3Sixty’s approach. They’ve focused a lot on how decisions are being made inside HR, IT, and finance orgs and how people are getting the most from their technology investments. Not that there isn’t value in quadrants or waves, or those incredible feature/functionality breakdowns I personally love consuming. But it is also a complicated, noisy time. As a former practitioner, I am empathetic about the decisions leaders have to make. Trying to make sense of that noise is important.

The last is infrastructure. While it’s boring, I know how hard it is to drum up new business, get contracts signed, follow up, invoice, design, QC, etc, etc, etc… I know people outsource parts of that and all I can say is that they are probably far smarter than my DIY approach. While I’ve appreciated the hard knocks education, I’ve also come to appreciate that infrastructure way more than I ever thought about when I was a W2.

What does that mean? You’ll see more of my research out in public instead of white labeled and behind another brand. You can brief me now and I have a pretty good process and expanded audience for market coverage. You’ll still get this newsletter. You’ll still see me writing for Reworked somewhere or on a random Rep Cap call, too. And if you’re still interested in some weird little project, you can still get me. Although, probably a little less of that.

If you’ll be at HR Tech, let me know. It would be great to catch up. And if you want to request a briefing, you can do that here.

University of Phoenix’s parent company agreed to buy Fuel50 earlier this month, and my first reaction was surprise.

Universities don’t usually buy HR tech. You probably knew that.

Then I thought about it some more and got a lot less shocked. Now I’m surprised that more colleges (and the companies that work alongside them) aren’t doing something similar.

I know Jo Mills and Anne Fulton, Fuel50’s co-founders, and I’ve worked with them on multiple initiatives. I admire both of them a lot, so factor that in.

Both employers and colleges are struggling to connect. Pearson and AWS heard from more than 2,700 people in six countries this spring. Only 28% of employers said universities are keeping pace with AI-driven change. And 78% of higher ed leaders said they were confident employer expectations were being met.

Different questions, so I won’t push it too far, but you’d also think those two numbers would be a lot closer together.

Adult learners aren’t exactly cheering either. The number of first-time students over 25 fell 15.5% from fall 2024 to fall 2025, according to National Student Clearinghouse data. Phoenix is built for working adults, and its third-quarter revenue was $271.8 million. That’s basically flat, which isn’t a decrease but it isn’t an increase.

I’m a big believer in higher ed. I’m just as big a believer in adults building their skills at any point in their careers (including going back to school), so those numbers bug me.

Employees are feeling the squeeze. AI is changing what their jobs ask for, and the ways to reskill, from degrees to bootcamps to whatever the employer offers, add up to a long list that’s tough to sort. Employers are trying to help. Strategic Education (the parent company of Strayer and Capella) said 34.7% of its U.S. higher education enrollment came through employer affiliations in the second quarter, up from 31.8% a year earlier.

A better connection between what employers need and all those options is a good thing. I’d like to see more of it.

So I get why Phoenix wants in.

My worry is the obvious one. Phoenix sells degrees, and every skills gap Fuel50 finds is a possible enrollment. If Fuel50 keeps doing what it was built for, helping employers see the skills they already have and move people into new roles, employees come out ahead. If it turns into a marketing channel for future Phoenix students, they don’t.

Mills wrote in her LinkedIn post about the deal that Fuel50 continuing under its own brand “was important to us.” I’m taking her at her word, though the real proof will take time.

The rest of my thinking, including what education companies have tried with workforce data before, is in my Reworked column. Give it a read and tell me if you get it or not.

What else is going on? A little change

I am experimenting with a new format. I’ll share some of my favorite pieces from voices I like as well as a more straightforward news round up. Let me know if you like it (or not).

Industry Voices

• The Best HR & People Analytics Articles of September 2026. David Green‘s monthly list is always a must read, but fill your coffee first.

• The Chatbot Talked Your Buyer Out of It. Rob Catalano is on fire, and this week’s pearl is a question to ask before your next AI contract: does the vendor make more money when your agents are inefficient?

• On AI, Maths, Proofs, Theory and Work. Thomas Otter admits he’s prone to long, convoluted Substack posts, then quotes several leading mathematicians on OpenAI and the Navier-Stokes proof. We love those posts too.

• The Future of Work Has a First-Day Problem. Jason Averbook asks CEOs whether they’d want their own child in the entry-level job AI has changed. Good questions, especially if you’re a parent.

• Notes from the Blank Page: What AI-Native Really Looks Like. Love the transparency in Jess Von Bank‘s note on where Now to Next is still stubbornly pre-AI and where the tools disappoint.

• Your “Best Practice,” According to Whom?. Robin Schooling compares “best practice” to the mattress tag you’re told not to remove: it sounds official, discourages questions, and few people can say who decided it. Always my big question here.

• Candidate Fraud Has Arrived. Here Is How HR Leaders Can Build a Defense.. Dan Schawbel says candidate verification takes buy-in from HR, IT, security, and legal well before any vendor demo, and this is going to come up more and more.

• Find the Person in Your Organisation Who’s Already Building the Culture You Want. I love this idea from Mervyn Dinnen, which starts with handwritten letters from a senior VP that an employee still keeps in a safe decades later.

• The Trust Layer: Why the Point Vendor Era Is Closing. Charles Bedard found identity and data assets in workforce compliance deals sold for 3.7x to 12.2x revenue, against 1.0x to 2.8x for per-check screening. It might seem counterintuitive, but there’s some there there.

• Road Report: RecFest USA, Nashville. Mike Wood gives people a taste of Nashville, from Juicebox’s capybaras and mechanical bull to the room applauding loudest for keeping interviews human.

• LinkedIn Is Great at Reaching a Network. It Was Never Built to Serve an Audience.. After about 80 editions, Paul Slater is moving his newsletter to Substack. His newsletter is a must read.

• AIQ x2: Pacing the Frontier, and Who’s Actually Auditing TA. Chris Havrilla and Bob Pulver made this biweekly live conversation official this week, taking on an ATS you can’t explain and what “human in the loop” needs to mean. This series is so good.

• 7 Rules for Talent Leadership in the Age of AI. Great rules from Kevin Wheeler, starting with “You Are Not a Recruiter Anymore”: the job is building the best talent system, not being the best recruiter.

• Seat Change: SaaS Isn’t Dead. But Everyone’s Trying to Figure Out the Next Pricing Model.. Give me the dad joke in the headline every day. Steve Smith says incumbents are giving AI away while startups like Ema bill for completed work instead of seats.

• Investors Are Cold on Talent Acquisition Right Now. BoomerangHR Just Raised $12.7M Anyway. Good info from George LaRocque, who sat down with the CEO to ask why a bet that your best next hire already worked for you is landing when most TA startups can’t get a meeting.

• Self-Leadership Is the Most-Requested Speaking Topic of 2026. Laurie Ruettimann defines it as the art and science of individual accountability and points to her LinkedIn Learning course. This is the only way to survive corporate.

• UNBOUND Recap: HubSpot’s Next Act Looks a Lot Like Work Tech. Alec Cherrin came back from HubSpot UNBOUND convinced some of it is worth turning on tomorrow, and less sure about the rest. Check the replay with Steve Smith and ask Alec your questions if you still have them.

• I’m Back! Let’s Talk About HR Tech and Measuring What Matters. It’s great to see Meghan M. Biro publishing again, with The Work Tech Pulse relaunched as a monthly LinkedIn newsletter.

• Blizzard Entertainment Workers Ratify Historic Video Game Contracts with CWA. Steve Hunt reads Blizzard’s AI bargaining language as a sign AI will likely drive more unionization in historically non-union jobs. I’m seeing many of the unions covering creative arts already trying to put AI restrictions in place.

• What Phenom’s Voice Screening Agent Looks Like in Production. Nicole Roberts talked to talent leaders at U.S. Venture and SASR Workforce Solutions, and the takeaway is that every applicant gets a real conversation right away while the hiring decision stays with a human. This technology should be good.

• Findem Studio Is Open to Everyone. Love this from Bennett Sung, who says finished talent acquisition work doesn’t mean the decision is made, just that you have something real to work with sooner.

• The Rep Is the Point. Brian Fink makes the case for writing: the blank page won’t let you fake understanding. Brian gets it.

• The WNBA Has It All. Leela Srinivasan backs it up with elite play, stars fans care about, and finally being able to watch it. A little off topic, but the WNBA has been great, and turns out you just need to pay your stars and get good basketball stories covered.

In The News

• Meta’s New Enterprise Push Starts With an Old Question: Will It Last? via Reworked

• Trump Administration Settles With Unions Over Shutdown RIF Challenge via Federal News Network

• Blizzard’s First Union Contract Sets Layoff Terms and AI Rules via HR Executive

• US Job Openings Fall to Fewest Since March, Layoffs Subdued via Staffing Industry Analysts

• California Expands Harassment Prevention Training to Address Hate Speech via National Law Review

• OSHA Urges Employers to Prepare for Opioid Overdoses at Work via SHRM

• Insurers Say AI Could Add Billions in Health Costs. Billing Companies Disagree. via HR Dive

• Anthropic’s AI Scenarios Put the Early-Career Talent Pipeline in Focus via HR Executive

• Fraud Hits Employer-Jobseeker Trust in Hiring, Report Finds via Human Resources Director

• Job Seekers Willing to Take Lower Pay to Get AI Training via ERE

  The AI Utilisation Gap: Why Employees Use AI More Deeply at Home Than at Work via Workplace Insight

• Agentic AI Early Adopters Have Failed, Pivoted, and Learned These 3 Lessons via Fortune

• Nearly Half of Workers Say They’re Exaggerating Their AI Use at Work via HR Brew

• More Independent Workers Go Full Time, Plan to Continue via Staffing Industry Analysts

• The Operational Fix That Could Jump-Start Innovation? A Week With No Meetings. via HR Dive

• Major HR System Hacked Again as Criminals Slip Past Summer Fixes via Human Resources Director

• Uber CEO Says ‘Really Demanding’ Work Culture Includes Expecting Employees to Answer Emails Over the Weekend: ‘Don’t Come Here If You Want to Coast’ via Fortune

• This Is What Your Highest Performers Really Want via Fast Company

Have a great rest of your week!

Sam Altman built his reputation partly on boldly predicting AI would gut the job market.

He told CNBC in 2019 that AI would “probably replace most of the jobs people do today.” In July 2025, he told Congress entire job categories would be “totally, totally gone.”

Ten months later, during a virtual interview at a Commonwealth Bank of Australia conference in Sydney, he said he doesn’t expect “the kind of jobs apocalypse that some of the companies in our space advocate or talk about.”

Some of the companies?

Hey, Sam, that was you.

This is the part of tech leadership I have zero patience for. The amnesia and lack of accountability for people who predict the future from positions of power. Altman goes from “totally, totally gone” in front of Congress to “that’s some other company’s rhetoric,” all inside a year.

What changed?

If you ask Microsoft’s Mustafa Suleyman, nothing. He thinks Altman’s initial prediction was right. He’s telling reporters that most white-collar work gets automated within the next 18 months.

Of course, he isn’t trying to line his company up for an almost inevitable record-breaking IPO.

But what’s really happening is really, really important. So who do we believe? We might get a lesson from the last time an industry swore a machine was about to make its jobs obsolete.

In the 1970s, ATMs were going to empty the teller line. Machines that dispense cash for free don’t need a human standing next to them. Lines at the bank would become a distant memory and bank tellers would fall the same way telephone operators did.

You didn’t have to be Sam Altman to make such an easy prediction. It was a widespread belief.

But it didn’t happen that way. Tellers per branch fell, sure, from 20 in 1988 to 13 by 2004, but cheaper branches meant banks opened more of them. Urban branch counts grew more than 40% over that stretch. ATMs didn’t stop growing either. There were 383,000 of them by 2004.

The teller job changed but it persisted. Then the smartphone showed up

Mobile banking did what the ATM never managed. You could manage money, make transfers, send wires, and do all kinds of complicated transactions from the device that lives in your hand.

I’ve been inside my own credit union maybe twice in the last five years. Teller employment went from about 332,000 in 2010 to 164,000 by 2022, and Bank of America alone closed roughly 40% of its branches between 2008 and 2025.

Even common sense is wrong sometimes.

Nobody at those banks was dodging IPO questions or lying to shareholders. They just watched the thing everybody was worried about while missing the thing that would actually make banking change for good.

So this isn’t really a piece about Sam Altman being full of it (even though he is). Post-IPO, his messaging will likely change again. The boy will cry wolf once again.

But it’s too easy to pick on him.

It’s about how hard it is for most people to see the real threat coming until it’s already inside the building, including people who have no reason to lie about it.

Watching the wrong door is a human problem. Never having to cop to it is a rich guy problem. Okay, maybe I’ll pick on him some more.

I get into the actual numbers on this in my Reworked column this week, including the CFO survey showing companies are cutting jobs faster than they can prove the AI paid for itself.

What else is going on this week?

• Your AI Adoption Gap Is Actually an Adaptability Gap. Jess Von Bank argues the AI adoption gap is really an adaptability gap, and layering AI onto old processes is “like hitting 1.5 on a podcast playback to listen to it faster.” This is the correct read.

• Acadian Ventures Inaugural Impact Report. Acadian’s first impact report counts 50-plus portfolio companies, $1.5 billion raised and more than 3,000 jobs created, and Jason Corsello and team are doing great work.

• Thinking, Fast and Hired. Brian Fink‘s argument is that every recruiting tool sold as AI is “functionally, a System 1 machine,” and the fix is structured, consistent scoring. I love his advice to buy the software, then buy the book.

• We Need Standards for AI in Hiring. David Manaster says recruiting lacks a standard for AI in hiring, so he drafted four, starting with a ban on auto-declines over criteria that weren’t stated upfront. He’s right, and while it’s tricky, that doesn’t mean it shouldn’t be done.

• The Governance Series So Far: AI Governance 8. If you haven’t been keeping up with John Sumser‘s AI governance series, here’s the TL;DR, including his line that “the machine’s recommendation is officially optional and socially mandatory.”

• Judgment Is the (OUR) Job Now. IBM found 71% of CHROs now call supervising and overriding AI the most essential skill, and Jason Averbook writes that 2026 proved the agent could do the work. I don’t even know if 2026 proved that, but agents can certainly make work.

• How to Stop Thinking (A Whimsical Way That Works). We definitely need an overthinkers anonymous, and Kat Kibben‘s version involves hiding tiny dinosaurs for strangers to find.

• HC Insider: September 2026. I don’t miss one of Kyle Forrest‘s newsletters, and this one has Deloitte’s new videos on HR jobs with AI agents in them.

• The €10 Billion Deal HR and Payroll Leaders Should Be Watching. Anita Lettink reads Silver Lake’s Cegid and Silae merger and its reported Workday talks as investors deciding the payroll-in-HR-or-finance debate doesn’t matter much anymore. I don’t like platforms leading the way, but that’s what happens when you don’t have your own vision.

• The Solution Premium Playbook: Why Software + Services Is Converging Now. Services firms are buying software companies at twice the rate software firms are buying services this year, according to Charles Bedard. I’m seeing this more too.

• What if There Was TSA PreCheck for Recruiting?. Doug Berg is trying to solve the tricky problem of candidate fraud by verifying licenses and certifications before someone applies. He says Match2’s “certify to apply” approach has produced 90%-plus reductions in fake applies.

• The Jobs Did Not Come Home........ I got introduced to a new term from Kristy McCann: “stability stacking,” Glassdoor’s name for layering skills, income and networks so no single job can knock you flat.

• Who’s Flying the Plane?. Robin Schooling worries about never-skilling and mis-skilling as AI takes over routine work. If we never learn the easy stuff, we might mislearn the hard stuff.

• The Fuel50 Acquisition: How Does This Make Sense?. University of Phoenix’s parent is paying about $31.5 million for Fuel50, roughly 1.5 times revenue by Steve Smith‘s math. I think it makes sense, and since it isn’t my money, I don’t care about the deal size.

• ZoomInfo’s New Recruiting Product Has One Job: Prove the Data Is Good. ICYMI, here’s my take on ZoomInfo’s Talent Autopilot, including research from Madeline Laurano on why recruiters don’t trust vendor data.

• The Future of Learning and Development. Smart write-up from Meg Bear on how L&D can help reimagine work instead of only leading AI enablement, part of what she and Amy Wilson have been pondering about agentic co-workers.

• Grumpy PSA for HR Tech Founders. “If you say you are doing recruitment selection with AI and no one on your team has serious industrial psychology chops you are building a zodiac.” Thomas Otter should just register the grumpy PSA trademark at this point (he’s also right).

• Women Applying AI on Womansplaining AI. Pam Boiros and Luda Kopeikina talked about Women Applying AI, the community they started, as it nears its first anniversary. Love the show’s title.

• The Role Work Plays in Recovery. September is National Recovery Month, and I love this conversation between Jeanne Achille and Heather Bussing on the role work plays in substance use and recovery.

• Business as Usual Was Never Stability: How to Lead When Disruption Is the Norm. Aaron Delgaty does tremendous work, including SparkEffect’s new Trust Report on disruption at work and the podcast interview that goes with it.

• The Frontline Worker Expectation Gap. 84% of frontline workers value clear pay insight but only 57% are highly satisfied with it, one of the good insights on the gap Pete Tiliakos pulls from Workday and DailyPay’s white paper.

• AI Adoption Is Driving Hiring, Though Mostly for Senior Roles. Stanford looked at 1.25 billion job postings and found senior employment at AI-adopting firms rose 6.7% over five years while junior employment fell 3%. More good news if you’ve got experience, but question marks remain for everyone else.

• Indeed CEO Says the Job Market Is Stuck in a ‘Vicious Cycle’ if 1,000 Qualified Candidates Can’t Get Hired: ‘Something’s Wrong’. Doesn’t Indeed play a role here though, though?

• HR Tech 2026 Announces 33 Contestants for Industry’s Most Popular Startup Competition. I’m looking forward to seeing the 33 startups George LaRocque and the Pitchfest team picked from 80-plus applicants for next month’s HR Tech.

• Workday Stock Rallies Following Reports of Ongoing ‘Go Private’ Financing Efforts. CNBC is hearing “a bit more optimism” about financing a Workday take-private, but nothing is confirmed yet. I’m starting to hit the “I’ll believe it when I see it” stage, but I’m also not a speculative investor.

• What the Apple-OpenAI Trade Secrets Fight Means for HR. With all the talk about AI, the Apple lawsuit against OpenAI is a reminder that getting the basics right is still critical.

• Only 58% of Employees Say Leaders Have Explained How AI Serves Company Goals. Culture Amp found 85% of employees say they’re encouraged to experiment with AI, but only 58% say leaders have explained how it serves company goals, and explainability matters here. Amy Lavoie adds that seven in ten feel more productive with AI while their workload scores haven’t moved.

• Video: The HCM Technology Utilization Gap: Why 87% of SMBs Aren’t Using Their Platforms to the Fullest. Dylan Teggart gives a good rundown of why only 13% of business leaders in 3Sixty’s research with Insperity feel they fully use their HR platforms, from weak implementations to poor support.

Have a great rest of your week!

ZoomInfo launched Talent Autopilot today, an AI agent built to reason across a database of 500 million professional profiles and surface recruiting candidates that keyword search would never catch.

For the last decade, every sourcing vendor's opening line has been a number, usually hundreds of millions of profiles. Every time, I ask the same question: Are they any good? Are they right?

Most vendors get defensive back. There’s bad data everywhere, they’ll tell you. Here’s a tool to sort through it. And, more recently: look at the AI.

That’s always the solution.

In my experience, recruiters’ actual complaints are rarely about volume. It’s saliency and accuracy. You don’t need another message that reads, “I noticed you’re a VP of Engineering at Acme. I’m hiring for a similar role at a company you’ve never heard of.”

ZoomInfo is really selling trust in its own data, not simply a smarter agent.

That might be an uphill climb from a company that told Wall Street its own AI story confused customers enough to cut its guidance this spring.

Recruiters Trust Their Data. Their Data Doesn’t Deserve It.

Madeline Laurano's Aptitude Research found that 77% of respondents would eliminate or downgrade a vendor that can't explain how its data is sourced and verified, in research ZoomInfo cited in its own launch announcement.

The rest of the numbers come from the leave-behind ZoomInfo handed out at its analyst briefing last week.

• 74% of recruiters say they're confident their candidate data is accurate

• 61% believe at least a quarter of their ATS records are already stale

• 62% say candidate information goes bad within three months

• 61% spend four or more hours a week manually fixing it by hand

Those are real problems that ZoomInfo believes it can solve.

What Talent Autopilot Actually Does

I’ve talked to ZoomInfo a handful of times over the years, out of personal interest and because the company has spent years trying to prove its data banks could do more than fuel sales and marketing lists.

Instead of matching on titles and keywords, the agent takes a hiring objective and reasons across skills, experience, career paths and company intelligence to surface candidates who wouldn’t turn up in a normal search. “They need technology that understands what they are trying to accomplish,” said Jon Bischke, general manager of ZoomInfo Talent Solutions, in the launch announcement.

Recruiters plug in the role, review what the agent finds, and refine it with plain-language feedback. Approved candidates get pushed into an ATS with duplicate detection, then reached through connected Gmail or Outlook accounts with personalized outreach and one automatic follow-up. More importantly, nothing goes out until a recruiter clicks send.

Beta customers ZoomInfo quoted in its own materials back that up. One construction-industry customer said trying it against LinkedIn side by side was “night and day” on candidate quality.

The consumption question that has plagued every SaaS company came up in the analyst briefing, too. When ZoomInfo priced Talent Autopilot on a per-job, capped-block model, HR-tech analyst Rebecca Wettemann of Valoir pushed back on calling it “consumption-based” at all, warning the word spooks buyers who spent the last year watching AI token bills rocket higher and higher.

I am less concerned about that, though. Recruiters have been paying for consumption forever. More jobs = more spend. But recruiters are also used to more granular controls and caps. It seems like they’ve listened to their buyer here.

ZoomInfo Just Had a Very Bad Spring

ZoomInfo cut its own 2026 revenue guidance earlier this year, from $1.247 billion-$1.267 billion down to $1.185 billion-$1.205 billion. CEO Henry Schuck told investors the company saw “a trend of AI and agentic confusion” in customer conversations that spring, one that paused purchasing decisions industry-wide. The stock dropped 32.78% the next day, from $6.04 a share to $4.06.

A class action, Tejeda v. ZoomInfo Technologies, is now working through federal court in Seattle over whether the company misled investors about how its AI products were resonating with customers.

Today, GTM trades around $3.88, down from a 52-week high of $12.31, according to real-time data from S&P Global. Its market cap is down about 70% over the past year. Even the average analyst price target, $4.60, sits above where the stock trades now.

Wall Street already thinks it’s underpriced, but buyers are still waiting for something long term to get excited about.

A launch like this might buy the company something else: proof to Wall Street that its AI story runs deeper than one bad quarter. But it’s still asking recruiters to trust a new AI agent with it’s nearly two-decade lead in data.

Delivering Results Is the Only Proof Left

In my view, the idea behind Talent Autopilot is sound. If you’ve hired anyone in the last five years, you’ve been burned by data that was out of date, incomplete, or stretched the truth because it was self-reported.

Verified, current, searchable data fixes a real problem.Whether it works is a different question.

ZoomInfo can only answer it by delivering a product that delivers on talent acquisition’s real pain points, not just presenting slides about it. The company has spent the year proving AI claims and customer experience don’t always match.

Now it’s betting its next product on the opposite being true.

We’ll know soon enough, the same way everyone else will: whether the candidates it finds are ones your team actually wants to hire.

ZoomInfo’s headquarters sits along one of my favorite walking routes in Vancouver. I’m selfishly rooting for them, because when ZoomInfo does well, my city is better. I like our city having jobs that aren’t based in Portland.

But I still want the receipts. My guess is we’ll see them soon enough.

That's great, it starts with an earthquake
Birds and snakes, an aeroplane, Lenny Bruce is not afraid

While the words of the R.E.M.’s immortal It’s the End of the World As We Know It goes through your head, Jacob Coxon is afraid.

Coxon is 27. He spent three years doing pretraining research at OpenAI and then Anthropic, and last week he quit. He told his coworkers on Slack, and then he told everybody, that without a change of course, AI has a real shot at ending us.

This is not, he wanted you to know, a marketing stunt.

The post hit 156 million views.

Two days later, the guy who actually runs Anthropic’s safety research agreed with him, on the record, in public: more than a 10% chance AI kills everyone within the decade. You know, the people getting paid the most to know better now think there’s a real chance the thing they’re building ends us.

Bill Gates picked this same week to remind everyone that AI will either be the greatest equalizer ever invented or the worst source of injustice, and that any model that can design novel molecules should probably be watched closely. His foundation is putting a billion dollars behind the equalizer version of that bet.

Cool, cool, cool.

Then, on Saturday, Anthropic’s CEO published 3,800 words asking the entire industry to slow down. He committed Anthropic to giving outside evaluators the same access as an employee. Sam Altman, Elon Musk and Demis Hassabis all said, sure, count us in.

Altman went further. He said OpenAI can’t currently deploy its most advanced model safely, because it hasn’t solved for whether the model will actually do what it’s told.

The four of them are also reportedly in talks to build a joint safety standards body.

Cool, cool, cool.

Here’s the part that irks me.

These four guys did more than anyone alive to build the race they’re now asking each other for permission to slow down in.

They set the pace.

They signed the compute contracts.

They shipped models on a breakneck release schedule, whether anyone understood what was inside them or not.

Slowing down always looks like losing rather than caution, though. So the fix is a group pact instead, because none of them trust the others enough to slow down on their own.

I’m not saying they’re wrong about the risk. I have genuinely no idea, and I don’t think they really do either. That’s the whole point of a probability estimate instead of a fact.

But I’ve got a wife and daughter. Groceries are going through the roof. There’s more than one war going on. The climate is already doing the thing they said it would do twenty years ago. The world was already on fire.

Extinction risk from a chatbot company isn’t getting the attention it deserves because we are tired. So very tired.

Affordability, not AI, is shaping up to be the actual top issue heading into the midterms. That tells you where most people’s attention already is. A Slack message from an engineer warning that the app that puts a top hat on your dog might kill us all doesn’t compute.

So no, I don’t spend a lot of my day wondering whether Claude or ChatGPT ends the species by 2036. God, I hope it doesn’t. Will it come before the Cascadia Earthquake?

I’ll be singing R.E.M., regardless.

The billionaires who built the thing they’re now afraid of would like the rest of us to be more scared, please, and preferably before Washington makes them stop on its own terms instead of theirs.

Maybe get your own house in order before you ask the rest of us to panic on your timeline.

Do your job.

Fix your shit before it goes to shit.

Please and thank you.

What else is going on this week?

• Graduating in AI Era Is Like a ‘Large Recession’ for Starting Pay. Graduates in the most AI-exposed majors saw their odds of landing a job drop 5 points and their starting pay fall 13% after ChatGPT arrived, comparable to graduating into a real recession. The white collar downturn for new grads keeps showing up.

• HR Must Set Boundaries as It Embraces AI Notetakers, Experts Caution. Some AI notetakers don’t even flag that they’re recording, which gets messy fast once you’re talking to candidates or employees across states with different consent laws. SHRM’s CHRO Jim Link put it well: if you’re using one of these right now, “I would want to have all of my ducks in a row.”

• AI Resumes Meet AI Recruiters: Dispatches From the Automated Hiring Cycle. Candidates use AI to write resumes that beat the AI screening the resumes, so companies buy more AI to catch it, and both sides end up with less useful information than before. As Greenhouse CEO Daniel Chait says, “It’s not you; it’s the system, and it stinks.”

• DHS Proposes Ending 60-Day Grace Period for H-1B Workers. The rule would strip the grace period that lets H-1B workers stay in the country and look for a new job after a layoff, dropping to a 10-day window in some cases. Immigration attorney Steven Brown points out the obvious problem: a company that cuts someone two years into a three-year approval leaves them with no cushion at all.

• Workday Has a New CMO. Sarah Kennedy Ellis, who spent the last decade running marketing at Google Cloud, Adobe Experience Cloud and Marketo, takes over Workday’s global marketing starting in October.

• Frontline Workers Want More Control, Flexibility. Most hourly workers say they’re proud of their jobs, but 69% still feel replaceable and unpredictable scheduling is their biggest stressor. Deputy CEO Silvija Martincevic says pride comes from the work itself, but feeling valued comes from how an employer actually treats you.

• ‘Maxed-Out’ Managers Need Backup: 4 Moves for HR. Korn Ferry’s survey of 16,000 workers found 42% say their company has thinned out its management ranks in the past year, and the ones still standing report more exhaustion than CEOs, senior leaders or anyone else on the org chart. “Some are watching senior leaders get buried in constant org change and deciding people leadership just looks like a headache,” says Ryan Starks of Rising Team.

• Business Leaders Plan to Continue AI Spend, Even if Bubble Bursts. Accenture found 52% of business leaders say they’ll keep investing in AI even if the bubble pops, with just 10% thinking there’s a real bubble to begin with. “AI investment and conviction holds,” says Accenture’s Karalee Close, though whether that conviction survives an actual bust is a different question.

• SHRM Rebrands Annual Conference as ‘Credentialing Academy’. SHRM is replacing its summer conference with “SHRM HR-X Live!,” which sorts attendees into four “dimensions” and hands out an HR-X credential at the end. Hmmm…

• This Is My Love Letter to HR. Heather Bussing makes as good a case as you’ll find, with ADP’s backing, that HR pros have always belonged at the center of business strategy, not on the sidelines of it.

• This Week in Work Tech, a Schedule Conflict Presents a Choice for Conference Goers. Two must-attend HR tech conferences are landing on the same dates this year, and Steve Smith breaks down whether to burn the frequent flier miles on Vegas or Paris, which is only a fair fight because Vegas has its own Eiffel Tower too.

• Good Morning From the West Coast. I joined the Purple Acorn HR Morning Show this week to talk about how people are managing, and mismanaging, AI generated work, in case you missed it live.

• In This Vintage Edition of the Punk Rock HR Podcast. A vintage edition of the Punk Rock HR Podcast resurfaces Laurie Ruettimann‘s memorial to Claudia Martinez Foster, who died on 9/11, along with her idea for a new kind of Labor Day built around remembering and service.

• Pivot Spotlight: Mary Grace Hennessy: Taking the Hard Fork. Lexy Martin profiles Mary Grace Hennessy, who at 54 walked away from decades of remote consulting work to take an office job in a new city, calling it the hardest fork she’s taken since her twenties.

• SS Live: AIQ x2 with Bob Pulver & Chris Havrilla. Chris Havrilla and Bob Pulver get into the growing gap between how fast AI is moving and how slowly governance is catching up, with a pointed question for talent acquisition: who’s actually auditing what’s happening inside the ATS, and why.

• I Meant to Take the Week Off. David Manaster had shoulder surgery and rebuilt ERE Pro in three days anyway, mostly by talking to Claude. The real news is the new AI tool trained only on ERE’s 425 hours of TA conference sessions.

• The Board Already Owns This. Companies are handing AI agents authority that used to belong to people, without separating what those systems can see from what they’re actually allowed to do, John Sumser argues.

• The Edge of Now Episode 6: AI Is Moving Faster Than We Can Govern It. The sharpest line from Jason Averbook‘s conversation with Jess Von Bank is the one that sticks: superintelligence isn’t some future threshold, it’s just the point where humans lose control of the technology.

• Intelligence Is Not the Same as Humanity. Jess also draws the line between what AI can compute and what makes people people: values, judgment, relationships, the stuff a model can mimic but not actually have. Thank you!

• Why Your Product Is Invisible to Buyers (Even If It’s Really Good). Eight things decide whether buyers pick you over a competitor before anyone even opens a feature comparison, says Gordon Johnson, starting with familiarity: buyers choose what they know, not what’s best.

• We Have a Trust Problem. Shannon Pritchett opens by joking she’d been promoted to CEO of ZoomInfo, just to point out how easy that lie would be to make searchable and permanent. The real target is recruiting’s data problem: new research finds 61% of teams believe a quarter of their ATS records are stale. That’s it?

• Got To vs. Had To: How Do You Describe Your Work?. Whether recruiters describe hard assignments as things they “got to” do or “had to” do is the reframe John Vlastelica says he’s actually listening for in an interview.

• People & Progress Special ...Addition.... Kristy McCann and Mike Wood take apart August’s jobs report state by state and find the good headline number, 4.1% unemployment, sitting next to a labor force participation rate that’s been sliding for months.

• The Ripple Effect of HR Leadership. The most meaningful parts of an HR career rarely show up on a title or job description, Tom Darrow writes, they show up later, in the people and ideas you helped move forward.

Have a great rest of your week!

Back in 2022, Qualtrics launched something called CrossXM and I wrote a column asking a pretty basic question: How much do you actually have to spend on employee experience before it shows up in customer experience?

I used my local dive bar as the example. Great customer experience, no fancy CX tools, and they didn’t even take credit cards until a few years before that. I still think that’s a fair question, and I still don’t know the answer.

Four years later, Qualtrics built something bigger around the same question. I’m not sure they answered it yet, either.

Yesterday the company held a stream out of Ryan Smith's training facility in Salt Lake City to announce XM Data & AI. Qualtrics builds fake customers to stand in for real ones. It guesses what one specific customer does next. Then an AI agent writes that action straight into whatever system holds it: a loyalty program, a booking tool, or whatever else is plugged in. Qualtrics' names for that: Simulate, Predict, Trusted Outcomes.

That’s the pitch, anyway.

Qualtrics CEO Jason Maynard says the word EX out loud once, which was loud enough for me. But his line, or any other mention of EX, never makes it into the press release, or into any of the three writeups I found.

What’s actually for sale today: a consulting package (an X-Loop Diagnostic that maps what Qualtrics calls experience loops) and a bundle of reputation tools.

The stuff that actually looks like employee experience work, an app that routes a task to whichever worker’s up next, a prompt on someone’s screen mid-shift, comes out in 2027. So does the agent writing that action into somebody else’s system.

Qualtrics already updated its own EX suite this year, back in March at X4. It offered AI that asks a sharper follow-up question on a survey, plus attrition flags built by mixing HR records with survey answers. Group-level stuff, across a whole team. Nothing here guesses at one employee the way Predict works on one customer. Qualtrics’s own support page still limits synthetic panels to US general-population consumers.

I find the money more interesting than the roadmap, honestly.

In May, Qualtrics closed a $6.75 billion deal to buy Press Ganey Forsta, forty years of healthcare data across more than 41,000 facilities. That’s patient data. Squarely the customer side.

And Medallia, the closest thing EX has to its own heavyweight, got a fresh $150 million from a Blackstone-led group in June, a markdown five years after Thoma Bravo bought it for $6.4 billion. Culture Amp cut 70 people that same June, about 9% of its staff, its third round of cuts in three years.

Not a great year to be an EX-only vendor, if you’re keeping score.

So just look at where the money’s actually going. It’s flowing to the side that touches customers directly. Employee data just rides along, picked up as whatever the CX product happens to need to see about the person doing the work, and the CX budget picks up the tab.

But none of it works without somebody actually doing the job in front of the customer, though. A model can guess what a guest wants. It still takes a person to hand over the free upgrade, eat the complaint, or smooth over the delay.

Who’s watching whether that interaction actually goes well? An employee is. It’s the one who gets the prompt on their screen mid-shift, in the same demo where Maynard says the word EX, the only time all day.

That’s EX.

The one thing missing from the whole pitch. Same question I had in 2022, just with better production values this time.

Jobs change. Roles change. Skills change. Work changes. Life changes. That’s sorta how the world works.

Job architecture in particular is going through a reckoning, though. AI is accelerating change to the point where comp teams can’t figure out how to price a job.

And it’s easy to blame AI for all of this. I know I love to blame AI like I blame my hair for not covering my head. But I think this is actually an evergreen problem that has only been accelerated by AI.

ADP Research and Stanford's Digital Economy Lab tracked about 7,000 IT workers from 2019 to 2025 using real payroll data. The researchers looked at 5 million job postings and 9 million payroll records across more than 600 employers. They then mapped the work into 25 standard task categories and ran the wage data against each one.

They also did some controls to figure out how much of it wasn’t due to age, gender, or company size.

Their key finding? The work changed but not in the way they expected.

PwC's 2026 Global AI Jobs Barometer, released around the same time as the ADP-Stanford research, is the perfect companion for explaining what changed. They found that roles split across two broad categories.

Professionalized roles use AI to sharpen expertise that’s already there, like a radiologist or a recruiter. Those roles are growing at twice the market rate and their pay is climbing 42% faster.

Democratized roles use AI to let someone without the expertise do the job anyway, such as an IT service manager, a medical secretary. Those roles are growing slower, and so is the pay attached to them.

Those changes work their way down to the entry-level, too. Workers with highly democratized roles are struggling but so too are the entry-level folks in professionalized roles. The latter are expected to show up with senior knowledge at a discount rate.

Not awesome.

But behind all of this is a truth: We’ve always lagged on job architecture. Things change, we work, and the reference documents get updated… eventually. AI hasn’t really transformed a thing that wasn’t already happening. It just added fuel to the fire.

There is a way forward for those who are interested. I wrote all about it on Reworked last week. I love diving into this stuff when given the opportunity.

What else is happening this week?

• America’s Jobs Scare Recedes. August came in at 162,000 jobs, more than triple what economists expected, and Courtenay Brown and Neil Irwin read it as the summer scare being seasonal noise. I’m not sure that’s right, since a three-month average of 71,000, up from 20,000, is a recovery from bad and not much more.

• Job Seekers Are Hiding Invisible AI Prompts Inside Their Resumes. The old white-text-on-white-background trick is back, this time as instructions to the AI screener. Duke’s Neil Gong found it in about 1% of 200,000 resumes on hireEZ, up sevenfold since 2024.

• AI Interviews Rolling Out for Some Federal Government Hires, Sources Say. The federal government is starting AI-run structured interviews through CodeSignal for the Tech Force program. Scott Kupor‘s OPM has already encouraged agencies to bring AI into hiring with a human still in the loop, and officials expect more of it. Good luck?

• Salesforce Leads $166M Investment in HiBob. $166 million led by Salesforce at a $3.2 billion valuation, and Ronni Zehavi‘s pitch is the bigger part: HiBob becomes the source of truth on roles, skills, and reporting lines for whatever AI agent happens to be asking.

• When Is a Protest in the Workplace Not a “Workplace” Protest? NLRB Backs Microsoft’s Termination of Employees Who Organized Gaza Demonstration. The Board’s Division of Advice found the on-campus protest wasn’t tied closely enough to working conditions to be protected, even with a months-old petition about Azure contracts behind it. A workplace demand now has to be the actual objective of a protest rather than an afterthought, which narrows the definition.

• The Application Is Not a Strategy. Brian Fink puts numbers on it: applicant-to-interview rates have fallen from around 15% in 2016 to 3% in 2024, and people who send more than 81 applications land offers less often than people who send fewer. The application was never the strategy, it’s just harder to pretend otherwise now.

• An Alien Mind: OpenAI’s Chief Scientist Jakub Pachocki Concerned About Astra, Should We Be?. Probably. Josh Bersin reads the post on OpenAI’s new Astra model as honest about alignment definitions that don’t account for values, and asks how you judge a model’s values when you don’t know what it was trained on.

• “No White After Labor Day” and Other Authorless Rules. Loved this one. Robin Schooling on the rules nobody explains: the $25 expense threshold from an incident in 2011, the no-meetings-before-9:30 rule that outlived the executive who wasn’t a morning person, all of them still standing because asking costs more than complying.

• Why People Resist Technological Innovation. Senior leaders have 20-plus years of watching implementations fail, middle managers fear anything added to a job that already runs long, and frontline employees assume change means worse work or none. Steve Hunt would have you answer each of those before the rollout, which is the right set of questions to be asking.

• Mediabistro Weekly Drop: The New Drawn Edition. I love what Matt Charney is doing at Mediabistro, and this one gets at what happens when accountants hold final cut over the people doing the work.

• Candidates Don’t Trust AI in Hiring. That’s Our Problem to Solve.. Matt Alder is an optimist about AI in hiring and still leads with the data against him: ThriveMap found 49% of UK job seekers think it’s unfair for employers to use AI in recruiting and 41% want no AI in the decision at all.

• Talent Management May Be Dead. But HiBob Just Moved to the Front of Whatever It’s Called Next. Steve Smith was drafting an obituary for the talent management category when the Salesforce news landed. His thesis is that the software is toast and the data underneath it is the bigger opportunity, and yep, this is big.

• Research Note: Engagedly/Energage. Srikant Chellappa sold Engagedly to Energage, which runs Top Workplaces, and asked for an analyst to look hard at the deal. Strong note from John Sumser, who obliged with the question UKG ran into after buying Great Place To Work: do I have to buy your software to win your award?

• The AI Boundary Most Professionals Haven’t Drawn, but Should. Danielle Farage surveyed 1,000 Gen Z professionals and found 67% deliberately protect certain skills from AI. Mervyn Dinnen writes up her three-zone test for any task (delegate, collaborate, or protect), and the protect zone is the missing one for most.

• You Can’t Spell Hormones Without HR. Retirement gets a webinar, a calculator, three reminder emails, and a rep from Fidelity, while a hormonal transition that will hit a large share of any workforce gets silence. I laughed, then nodded, at Madison Butler‘s take.

• Hold Still While I Talent Advise You. TA leaders keep telling recruiters to become talent advisors, and John Vlastelica‘s point is that hiring managers often have no idea what to do with one. You can’t order off a menu you’ve never seen.

• A Way Too Early HR Tech Preview. Never too early, and Mike Wood expects agents in every booth this fall, with Madeline Laurano‘s judging showing only 12% of companies reporting agentic AI use and nearly every product submitted claiming it.

• Making Claude and Perplexity Part of My Team. Good to see Jonathan Goodman writing out in the wild. On the first engagement under his new business he gave Claude and Perplexity four named roles with explicit things each wasn’t allowed to do, then fired a fifth when its summaries flattened three executives’ distinct positions on one word into a single view.

• Cathy Engelbert Is Retiring as WNBA Commissioner. Leela Srinivasan on the WNBA commissioner stepping down at year end after seven years that, on valuations, media rights, and pay, put the league ahead of every other women’s league in North America. A change in leadership is a change in opportunity, and what got the W here won’t be what gets it to the next level.

• How to Build an AI Development Team That Can Deliver at Scale. My piece for X-Team’s Keep Moving Forward magazine, and you should read it anyway. Governance and domain expertise came up early and often, alongside architecture, when I asked five people who build these teams.

• Are We Watching the Economics of Work Start to Shift?. A good August recap that looks across hiring, layoffs, pay, benefits, and who’s participating, and asks whether the separate headlines add up to something bigger, from Mel Plett and Francesca Ranieri.

• Superpowered Launches in October. A first book on what AI does to individuals, teams, and organizations, written with Josh Bersin and out next month. Congrats to Kathi Enderes.

• Introducing Inspiring Place to Work. Matt Manners is turning the award into an identity: from 2027 every organization that meets the standard can call itself an Inspiring Place to Work, and 2026 winners get the badge now. That’s a big move for a program that has been running since 2015.

• The Meg & Amy Show Is 31 Followers Short of 1,000. Keep listening to Meg Bear and Amy Wilson, whose premise is that sensemaking is a team sport and collective intelligence beats individual expertise in times of change.

• The Global Talent Strategy and Intelligence Conference, Amsterdam, September 28 to 30. Jess Von Bank is emceeing Intelligence Group’s gathering at the H’ART Museum, with talent leaders from Philips, Booking.com, NATO, and Microsoft on the program. And yes Jess, you can have the pumpkins out now, as long as they aren’t real ones.

• The HR Morning Show Hits Episode 100. Evan White marks episode 100 of the HR Morning Show, featuring a baseline standard for ending candidate ghosting, a former USWNT player on high-performing teams, and the producer’s last turn in the chair. Congrats to the whole crew.

Have a great rest of your week!

Less than two decades removed from a devastating Civil War, the U.S. was in the process of becoming a burgeoning industrial power blessed by vast natural resources at its disposal.

While Andrew Carnegie and other powerful industrialists of the time get the majority share of historical credit for this movement, it was built off the literal backs of working class people. Employees regularly put in 12-hour days, six or seven days a week, with no legal protections whatsoever.

Get hurt on the job? Tough luck. Die on the job? Really tough luck. Hope your family has a good support system.

Peter J. McGuire and Matthew Maguire each are credited by some with proposing a workers’ holiday in New York City in 1882, and which of them actually originated it is still disputed. What isn’t disputed is that the first workers’ holiday in New York City ran this week in 1882.

Five years later, Oregon became the first state to make it an official holiday. Federal recognition came in 1894, driven in part by the Pullman Strike when 13 workers died during demonstrations.

Since Federal recognition, Labor Day has been a quieter cousin of May Day (President Grover Cleveland deliberately chose the former because it was too closely associated with socialism).

In the more than a century since Labor Day became a national affair, things have objectively improved for the American worker. And while it has been a positive for the average employee, I don’t think anyone would dispute that having a protected and empowered workforce has been pretty good for businesses as well.

While upward progress looks consistent in eyes of history, reality had it’s share of ups, downs, and sideways for the American worker as they asserted themselves. The workforce got more diverse, yet protections came later for those who don’t look like, well, guys like me.

And if it feels like I’ve been telegraphing the AI point (get it? 1880s? no?), I would say the problem started much earlier.

Nearly a century after those first Labor Day movements, the connection between the success of organizations in producing more and the workers getting their rewards from that production have grown more and more disconnected.

AI, I would argue, is simply the latest trend that will widen the gap between productivity and median wages.

It’s not going away any time soon. The Indeed Hiring Lab AI-postings tracker shows the share of U.S. job postings mentioning AI terms rising from 1.9% in January 2024 to 6.3% as of late July 2026.

A decade or more ago, white collar employees could look at workers’ rights movements as primarily a blue collar or frontline worker concern. Being detached was easy. Sure, technology continued to progress. But could technology really write a book, do data analysis, or create complex software.

Turns out it could.

While the true impact on jobs is still being figured out, the new labor movement is taking shape. Some are adjusting to new technologies while advocating for fairer treatment and reduced hours. Some reject automation and choose to execute work without the help of AI. Other novel solutions, like universal basic income, are being considered by a diverse group of people (including 11 Nobel Prize Winners).

Even if you have a healthy detachment from work, you can certainly see the value of labor as a concept and why workers deserve a share of it when things improve. And even billionaires, who seem more detached from reality by the day, can see the value of labor (even if they personally do not go out of their way to do things within their power to appreciate it).

After all, if AI takes all of our jobs, who is going to buy the stuff that they want to sell you?

Do we need a fresh view on Labor Day or are we content just rolling out the old charcoal grill, burning our favorite tube meat, and drinking the finest American domestic swill while we look ahead to cooler days and a long winter? It shouldn’t seem too anti-American that our collective success isn’t driven by a small group of very rich folks and that more of us are dependent on each other doing well too.

What else is going on this week?

• Bootstraps Sneak Peek & BIG Conference Kickoff. A Portland preview of “Bootstraps,” a decade-in-the-making documentary on what UBI actually looks like for the people who lived it, kicks off the BIG Conference on September 16. I’ll be there, and so will Laurie Ruettimann, who’s hosting.

• The Best HR & People Analytics Articles of August 2026. My own column on Uber’s AI budget made David Green‘s five-piece stack on why token spend isn’t the same as business value, one entry in his usual pile of dozens of links.

• Hugging Face Hack Could Indicate Cultural Issues at OpenAI. OpenAI’s postmortem on the Hugging Face hack documents employees flagging a rogue AI message board twice before the attack, but never asks why leadership let it slide both times. Zvi Mowshowitz‘s answer: the company’s safety culture is either missing or too weak to matter.

• I Let Grok Bot Apply to Software Engineering Jobs While I Was Sleeping. Bobby Hall Jr scheduled Grok Bot to search, score, and auto-apply to software engineering jobs every four hours while he slept, capped at ten applications so it wouldn’t chase quantity over fit.

• Has AI Become the Solution and the Problem in the Hiring Process?. 73% of HR pros say they’ve now seen fabricated or misleading candidate information, which is Mikaela Cohen‘s version of AI cutting both ways in hiring.

• It’s the Buyout, Stupid: What Workday’s Earnings Actually Tell Us. Steve Smith gave us his read on Workday beating every number and still dropping 5% the day Salesforce and ServiceNow popped: investors weren’t grading the quarter, they were grading the unconfirmed Silver Lake buyout rumor sitting on top of it.

• AI Governance 5: Aderit.ai Wrapup. Five posts into his AI governance series, John Sumser lands on the plainest description yet of why HR has no real system of record: every function keeps its own version of the data, and nobody reconciles them.

• Beyond the Spreadsheet Summit. Anne Fulton opens Fuel50’s Beyond the Spreadsheet Summit arguing workforce transformation is a decision-making problem before it’s a technology one.

• Deloitte to Pay $21.5M to Settle Claims Its DEI Programs Violated Federal Civil Rights Law. Deloitte will pay $21.5 million to settle a suit over DEI hiring goals tied to race and sex, with the whistleblower money going to the American Alliance for Equal Rights (helluva name for an anti-DEI org).

• AI and the Writing Profession. Michelle Rafter shared a follow-up survey on AI and the writing profession, and anyone who fills it out gets the report when it publishes later this year.

• The Real AI Advantage Isn’t Speed of Adoption. It’s Workforce Readiness.. Nearly 40% of employees say they’re concerned about AI, and Katie Carroll flags the worse number underneath it: close to half say their employer has left them to figure it out alone.

• HR in 2030: Half the Headcount and Three Times the Impact. 13% of HR roles are routine enough that AI just absorbs them, another 60% get elevated by it, and Kathi Enderes is already running workshops built around that split.

• The Open Sourcer Folder. Amybeth Quinn keeps a public folder of every deck and resource she’s built for talent sourcing conferences, and the latest addition covers how to show your worth as a sourcer.

• Everything You’ve Ever Typed at Work Just Became an Asset. A nice companion piece to my take on Google trying to buy employee data, if your idea of nice is Sharon O’Dea telling you to really watch what you write at work.

• Are Your People Silent and Stressed? Work-Life Integration Can Help. Work-life pressure ranked among the top concerns in 32 of 47 countries surveyed, and Donald Thompson‘s read is that it’s a manager-relationship problem before it’s a workload problem.

• Why AI Is Making Onboarding the Most Important HR Investment of the Decade. Only 12% of employees say their company nails onboarding, and Dan Schawbel‘s point is that AI has made it worse: new hires can no longer learn the unwritten rules from colleagues who are themselves still figuring out the new workflows.

• Indeed Wants Your Recruiting Budget, Not Just Your Ad Budget. David Manaster digs into Indeed’s earnings call: 30% revenue growth even as job postings fell 4%, and now a direct pitch to CFOs to cut recruiters and agencies instead of buying more job posts.

• The Superpower of Attention. Digital workers switch apps roughly 1,200 times a day, and Brian Fink cites research putting the refocus cost of a real interruption at up to 23 minutes, nearly five working weeks a year gone to reorienting. His fix: close the tabs.

• From Headcount to Capability: Rebuilding Talent Acquisition for the Disaggregated Firm. Large corporations exist because coordinating work through the open market used to be expensive, and Kevin Wheeler argues AI has eroded that cost enough that talent acquisition needs rebuilding around a small core and a flexible outside network.

• The Edge of Now Episode 5: AI Should Come for Your Org Chart. But Have You Earned the Move?. Jason Averbook and Jess Von Bank open their new podcast arguing that eight executives leaving Expedia in an “AI-driven shakeup” was actually a redesign, not a cut.

• Are You Operating Like a Contractor or a Consultant, or Something Else?. John Vlastelica has watched recruiters get called contractors, then consultants, now architects, and argues AI is eating exactly the contractor-shaped work: sourcing, outreach, scheduling, and reports. What’s left is judgment, and judgment is what justifies the salary.

• Smart Glasses Are Back! Time to Grade My Predictions.. Grading her own 2013 predictions on smart glasses, Anita Lettink found Meta has sold eight million pairs of exactly the tech she warned about. Her advice hasn’t changed: get a policy in place for what she calls the “pervert glasses” walking into meetings.

• How to Respond to an AI Bot Interview. Kat Kibben‘s fix for AI interview bots is blunt: over-explain everything, because a human can ask a follow-up and a machine can’t.

• Where Hiring Is Heading, in Five Guides. Matt Alder distilled 800-plus podcast episodes into five topic guides on where hiring is actually heading, from AI adoption to what employer branding means once candidates ask an LLM about you first.

• Back in My Day. Robin Schooling‘s take on “back in my day” workplace nostalgia leans on a psychologist who’s studied it for decades: the feeling isn’t really about the landline or the cheese in a can, it’s proof the person you were and the person you are now are still the same one. I’ll take that.

Have a great rest of your week!

In the early 1990s, Mark Turmell was working as a game developer at Midway Games in Chicago. A Michigan native and lifelong Pistons fan, Turmell was working in the heart of the beast. If you know just one thing about 90s basketball, you probably know a name and his team:

Michael Jordan of the Chicago Bulls.

The Pistons were bitter rivals of the Bulls. They won championships a year apart from one another. They also both beat my beloved Trail Blazers so they can both take a hike as far as I’m concerned.

During that run up to historic decade, an arcade classic was in the works. NBA Jam was one of the first games to broadly license teams and players from the association. It was fun for casual arcade goers and home console players alike. A trip to Wunderland with a bag full of nickels would give you hours of entertainment back then.

After it was clear that the Bulls were a force to be reckoned with, Turmell decided to get revenge on the Pistons’ nemesis in the game play. The trick? He made the Bulls worse on last minute shots against just the Pistons. You can watch the interview below (starting at about 19:40).

The mechanism was subtle. It wasn’t a hardcoded miss like most people might assume. He just tipped the scale to make them shoot, on average, much worse than they had been shooting all game. It looked like a choke job.

It wasn’t discovered until 2008, 15 years after the game was released. And we only heard about it because Turmell told us about it.

A little harmless fun driven by a rivalry, right? I think so.

There is a lesson here, though. The original NBA Jam game was about 10Mb in size. That was a lot of code back then, but I’ve hit web pages that are bigger than that today. Someone could’ve looked for the bias and found it. And yet, for more than a decade, people couldn’t explain why Scottie Pippen kept missing buzzer beaters.

Do I think anything that malicious is happening in the AI solutions that hit the workplace? It seems unlikely (but not impossible).

What’s more likely is the system’s thumb being on the scale in some unknown way. While intentionality isn’t a good legal defense for doing the wrong thing for the right reason (or so I am told), it’s still bad when it happens and it’s really bad when it happens in a black box that very few people can understand.

The system we use to hire, develop, and manage people are much more complicated than a now 30+ year old game. Throw some non-deterministic AI into the mix and that complexity multiplies. Just ask Derek Mobley and Workday as they continue to hash out discovery and evidence.

For workforce leaders, you don’t need to be a developer to understand it all but you do have to know what you know and what you don’t. You also have to grapple with the risk of unknown unknowns as one Secretary of Defense put it. That might make the choice less about mitigating risk and more about avoiding certain tools altogether.

At the very least, you don’t want your employees and the very competent employee-side counsels out there to figure out on their own why they are missing their shots, basketball or otherwise.

What else is happening this week

• Nearly One-Third of Companies Haven’t Deployed Any AI Tech in HR Yet, Study Finds. Only 31% of companies have deployed zero AI tools in HR, per Adam DeRose’s reporting on a new study. Budgets keep climbing anyway.

• My Radiator Hose Blew in the Middle of Nowhere. Mike Wood’s family broke down on the Delmarva Peninsula, and the strangers who stopped, a grandfather and grandson with a VCR bolted into their van, fixed it for $100 and a thank you. He judged them by the van first.

• How Technology Shapes HR, Not Just Enables It. HR’s tools and its culture shape each other. The keyword-matching ATS everyone’s used for 25 years came from what 1990s databases could compute, and AI-native platforms are about to do the same thing in reverse.

• Humanity Will Not Design Itself Into the System. Banning phones in class or laptops in meetings is easier than agreeing on what they’re for, and Jess Von Bank’s point is that the same logic is coming for AI.

• You’re Not Going to Like What Comes After Marketing. Steve Smith on G2 selling off three review sites and keeping the one AI cites, while Reddit lost 86% of its ChatGPT citation share in four days with no explanation. The “independent” evidence AI leans on is consolidating fast.

• What If AI Costs More Than the Employee? Andrew Spence’s breakdown: a human costs €56.69 per task, an AI agent €88.26 once you count review and rework. Leaders aren’t ready for that conversation yet, but it’s coming.

• YTD: What Actually Happened to Multiples in 2026. The SaaS Capital Index dropped from 5.58x to 3.82x EV/ARR in seven months, per Charles Bedard’s tracking. Airtable sold to Bending Spoons at 2.7x ARR, an 89% markdown from its 2021 peak.

• AI Scales Everything, Including Bad Management. Meg Bear’s point: AI doesn’t fix bad management, it just runs the dysfunction faster.

• The How #30: How Do You Know What You’re Seeing Is Valid?. Wikipedia’s accuracy sits somewhere between 80% and 98%, and every source is cited at the bottom of the page. Kate Achille tells us AI can’t say the same: reputable sites block it, so it keeps recycling the same handful of sources.

• Powerpoint Dollars and Productivity. John Sumser’s first computer was a 1981 IBM PC, and he used it to calculate $7.5 million in “savings” that actually bought each worker five minutes a day. Same math showing up in LLM productivity claims now.

• Unemployment Is Lying to You. Unemployment improved last month because 720,000 people left the labor force, not because they found jobs, says Brian Fink. Participation is already at a 50-year low.

• Who’s Accountable When AI Blurs HR and IT?. Mark Whittle’s read: CHROs and CIOs will have to co-own AI ethics and skills-based talent management, because none of it fits neatly in either function anymore.

• Nobody Wants to Own Candidate Fraud. Kyle Lagunas’s new report on candidate fraud makes one point: nobody owns it. “When nobody owns it, you can’t create accountability,” he says. H/T David Manaster.

• Feds to Propose H-1B Fee That Would Stack With Controversial $100K Payment. The proposed $103,265 H-1B fee stacks on top of the existing $100,000 fee. Ogletree Deakins’ Caroline Tang is already calling the number a likely target for litigation.

• OFCCP Dismantles Decades of Federal Contractor Affirmative Action Requirements. Sheila Abron and Jennifer Sandberg walk through the OFCCP’s three new rules, hitting 118,000 federal contractors and rescinding affirmative action requirements that have stood since 1965.

• AI Emotion Tracking Is Taking Employee Monitoring Into Its Most Personal Territory Yet. Korn Ferry’s Dennis Deans on emotion-tracking AI: an employee who looks anxious might just be worried about their kid, not underperforming, and the algorithm can’t tell the difference.

• How I Built an AI Writing Partner. Paul Slater keeps AI out of the actual writing, using it only for research and a final quality check. Drafting with AI and editing after, he argues, wrecks your own voice.

• Another Tough Year Ahead for Employer Health Costs. Employer health costs are projected to rise 9.5% in 2027, past $19,000 per employee, per Axios’ Adriel Bettelheim.

• The End of Company Culture as We Knew It. The foosball table was never real culture, Holly Grogan argues, just a symbol companies used instead of doing the work. Losing the symbol might force the real thing to show up.

• Happy Employee Referral Day. September 1 is now Employee Referral Day, one more manufactured holiday for HR tech’s calendar.

• Who’s Allowed to Help? Steve Hunt flagged Shonna Waters’s research on why disclosing AI help costs more trust than disclosing human help, even when AI does the better job. Her line: sounding like AI isn’t a knock on your intelligence, it’s a claim that the help you used isn’t seen as legitimate.

• Across 128,000 Prompts Analyzed, 97% of Students Used the AI Tool the Way It Was Meant to Be Used. Kristy McCann’s numbers: Pearson’s AI tool got used correctly 97% of the time across 128,000 prompts. Women’s share of technical roles at AI companies fell from 23% to 19% since 2021.

• The Generative AI Learning Penalty. Toby Culshaw shared research on Chinese students using AI for homework: they finished faster and scored higher, then did worse on the exam that actually tested what they’d learned.

Have a great rest of your week!

Editor’s note: Kristy McCann is a CEO, Founder, and CHRO who has spent 25+ years at the intersection of people, culture, and technology, building companies, transforming enterprises, and coaching leaders at every stage. We decided to go back and forth on a handful of deals that we think are about the same core issues.

Lance Haun — Three deals in seven days and they all three put a price on the same thing.

Lattice paid an undisclosed sum for Pando and gave the workforce record away for free. Silver Lake is reportedly in talks to take Workday private, no price officially reported. Google bid ten million dollars for a dead airline’s email.

Think these are all different? They’re not.

Kristy emailed me the morning the Pando news broke. Her read was that the M&A market is an employee TAM play and a data play, and that the transaction is the story. I wrote back that she had it backwards.

Kristy, great minds think alike, as long as they agree with me.

My position is that who owns these companies is the least interesting thing about them. Follow the money that far down the rabbit hole and you’ll miss the decision that was printed on the pricing page a year earlier.

She goes first. The deal she cares about is the one that almost nobody covered.

The Lattice MCP permissions

Kristy McCann — Start with Pando for a second and then leave it, because the deal is the least interesting thing here although interesting.

Fifteen people. No fundraise since 2022. Barbra Gago walks in as chief marketing and strategy officer. Marketing. Nobody buys fifteen people and a four-year-old cap table for the revenue, and Lance can run whatever multiple he likes on it. It won’t come out to a number that works as a purchase. M&A rollup time shall we say!

It works as the third move in a sequence. March, Mandala, and the coaching data. June, Lattice MCP. Tuesday, Pando.

The middle one is the whole thing and almost nobody covered it.

I read the setup documentation, because that is where companies stop marketing. When a user connects Lattice to Claude, here is what they approve. View employee information. View feedback. View goals. View competencies and career tracks. Then view and write meetings. View and write performance reviews. View and write updates.

Three of the seven are write.

So a manager opens Claude or ChatGPT, pulls a direct report’s review history, peer feedback, goals and competency track into the chat window, drafts the review there, and submits it. Never opens Lattice. Lattice markets the part where you blend it with Salesforce and Notion and Jira on the way through.

They will tell you, correctly, that permissions carry over. That manager could always see that review. True, and beside the point. The permission boundary held. The building boundary did not. That review used to sit inside one vendor’s walls with one set of terms on it. Now it is context inside a general purpose model, and the audit trail for how it got there is an OAuth consent screen somebody clicked in June. CHRO’s - something to really think about here.

Two things belong on the record, because the overstated version of this is how the true version gets thrown in the bin. An owner can turn the writes off. Lattice’s own instructions say you can set the connector to read only and prevent any writes across the organization. Default on, switchable off.

And Lattice disclosed all of it. The scopes are published. OpenAI is named on the subprocessor list, updated August 13, under a heading that says optional and opt-in. Lattice states OpenAI does not train on customer data and that it runs zero data retention. Nothing is hidden.

Which is exactly my point. It was never hidden and nobody read it. Fine print exists so that you don’t read it

So here is my question for Lance, since he thinks I have this backwards. Lattice is not selling performance software anymore. It is running a distribution strategy for a data asset and it published the pipe. The employee TAM was never the seat count. It is every record those seats ever produced, and Lattice just made that record portable into a general purpose model on the strength of a consent screen.

Tell me that isn’t a TAM play - as TAM Is the data or every employee

Why per-seat pricing opened the door

LH — It isn’t. She’s got the mechanism right and the motive backwards. The record didn’t get more valuable. The seat got less valuable.

Lattice sells seats. So does Workday, and the market spent eighteen months deciding seats are the problem. Workday was down about 15% on the year and more than 40% off its 2024 peak before the Silver Lake story broke. If agents do the work, you buy fewer seats. Per-employee pricing is a tax on a headcount number that stopped going up.

A vendor who thinks the record will hold his customers keeps the record in the building. You open it up when you aren’t sure the seat is going to be there in three years.

So I’ll leave the TAM correction where she put it. If this is a play for the total employee base, the employee base is the part in doubt. It’s why the stock got cheap enough for anyone to call and ask about it.

One more thing on her scope list. She was careful here and I’m going to be less careful.

If a manager can pull an employee’s file into a chat window, the file is in the chat window. Reviews, comp band, whatever else is on the profile. Tell me I’m right.

RIP Lattice’s payroll business — June 2024 to March 2026

KM — No. Stop there, because that’s the error everybody makes with this and it will get the whole argument thrown out.

The MCP server runs at manager and individual contributor level only. Administrative capability is not exposed, and Lattice says that applies to everyone, including a super admin who can see the entire directory through the admin interface. Data that lives only in that interface does not travel through this pipe. Performance context goes. The payroll file does not.

I’m being pedantic on purpose, because the payroll file is its own story and it is worse than the one Lance just tried to hand me.

Lattice announced payroll in June 2024, built with Gusto. HRIS went generally available that October. The help center page listing required payroll fields called for Social Security number, birthdate and permanent home address, along with base pay, classification, start date and work location. Ethnicity sits on the employee profile as its own field. That is the file that opens a bank account in somebody’s name.

That page now sits behind a Lattice sign-in wall. I read it when it was public. You can’t.

In October 2025, Lattice was still marketing itself as the all-in-one people platform with a modern HRIS, payroll and career development tools. About a month later, customers were told the HRIS and payroll were going away. Payroll ended March 31, 2026. HRIS ends July 31. I could not find a single public announcement of any of it. Not a press release, not a blog post. ADP built a migration page for stranded Lattice customers and even ADP hedged, writing that observers had speculated Lattice may be shifting focus and that ADP had not independently verified Lattice’s product plans.

Twenty months from launch to shutdown on the most sensitive file a company keeps.

Then read the retention page:

Your Lattice account and all data stored in your account will be deleted automatically 180 days after contract expiration or termination. Note: For Lattice HRIS Customers, this rentention [sic] period does not apply. All data will be deleted on the 6th of August.

The customers whose Social Security numbers were in the system got carved out of the standard policy and handed a fixed date instead. The page does not say which year. It also misspells retention, which tells you how much attention the sentence got before it went live. And what about compensation - hmmm

I’m not building a headline out of a typo. This is the retention notice for payroll-grade PII and it doesn’t carry a year. If you were an HRIS customer, go read it, then ask them in writing what was deleted, when, from which systems, and whether that includes backups.

And the paperwork never caught up to the product. Lattice’s Data Processing Addendum is dated April 3, 2024, before payroll and before the HRIS shipped. It enumerates the categories of personal data Lattice processes. Compensation and benefits are in there. Race and ethnicity are in there, in the special category paragraph. Social Security numbers are not. Dates of birth are not. Home addresses are not. The catch-all covers it, because “may include, but are not limited to” does exactly the work a lawyer put it there to do.

Now take Lance’s own argument and follow it one step further than he wants to.

He says the pricing model killed that product. Fine. Who sets the pricing model? Who decided a payroll business with SSNs in it wasn’t worth carrying another year? That is not a spreadsheet making a decision. That is an owner, and the reason I keep asking who owns these companies is that the answer is upstream of every roadmap either of us gets briefed on.

Oracle, the SaaS discount and the rate curve

LH — Here’s where we split, and I’ve got the really unpopular take.

Oracle is borrowing about forty billion dollars to build AI infrastructure and laying off thousands of people to cover it. No sponsor or take-private. No sovereign wealth fund on the cap table, either. A public board and a quarterly call drives everything.

Leading a public software company is now demonstrably worse than a PE or venture backed one.

Enterprise software now trades at a discount to the S&P 500. That never happens. Multi-quarter declines wreck equity as a retention tool, and companies under that pressure do hasty things. Surveyed financial executives said, 78% of them, that they’d sacrifice long-term value to smooth earnings. Fifty-five percent would skip a good long-term project rather than miss a quarter. That’s a public company problem and Oracle has it.

Lattice doesn’t. It isn’t public and it isn’t state sponsor-owned. It’s venture-backed, and it killed a payroll product with Social Security numbers in it inside twenty months anyway. Nobody needed a leveraged buyout to make that call.

On timing, everybody tells me these deals were waiting on the Fed. The Fed moved. The effective federal funds rate was 5.33% in January 2024 and it’s 3.63% this week. That’s 170 basis points.

But the ten-year Treasury went up over the same stretch, 3.95% to 4.71%, near its high for the period.

Nobody underwriting a seven-year hold prices off fed funds. Thoma Bravo’s Dayforce financing was a seven-year facility. So the investor waiting for debt to get cheap has been watching the wrong number, and his number got worse. The money never got as cheap as before. At some point everybody quit waiting on it.

A question back, then. One vendor is owned by a sponsor with a seven-year hold and a plan. The other answers to a market that just told 78% of its finance chiefs to protect the quarter.

Which contract would you rather be holding? Which one would you rather lead?

The PE exit backlog and Dayforce

KM — Neither, and that’s the answer, not a dodge.

I was in the room for the inflate and the deflate. I wrote the integration plan and the reduction plan for the same company inside eighteen months. Lance is describing a sponsor with a plan. I’m telling you what the plan is when the exit doesn’t exist.

Private equity is sitting on 33,575 companies it hasn’t been able to sell. Ten years ago that number was under sixteen thousand. At the end of last year it was 32,451, and it keeps climbing. US private equity returned 6.4% annualized from mid-2022 through this spring against 15.2% for the S&P.

They cannot sell what they already own. So why buy something enormous? Because the lot is full. This is a used car business. Buy the depreciating asset, detail it, flip it, and when nobody is buying, you buy a headline instead. It resets the story with your investors and buys five more years.

Lance says they’re moving now because waiting stopped paying. I say they’re moving because they can’t sell. Both are true and neither has anything to do with a product roadmap.

Where he loses me is the idea that the sponsor’s seven-year hold is the safer contract. Thoma Bravo closed Dayforce in February at $12.3 billion, a 32% premium, with a minority investment from a subsidiary of the Abu Dhabi Investment Authority. Sovereign money inside a US payroll platform, closed this year, and almost nobody wrote about that part. Reuters says Silver Lake has been in talks on Workday for months and may bring in additional investors. It has named none.

So I’ll ask rather than assert. You’ve seen who turned up on the last deal that size. Whose money would you like sitting inside the system that runs your payroll?

Oracle’s board answers to a quarterly call, which is bad. A sponsor answers to a fund with a clock on it and limited partners you will never see named. And the debt never services itself. It comes out of your renewal.

The Spirit Airlines data auction connection

LH — I’ve been saying the whole way through that the transaction doesn’t matter and the pricing model does.

I’ll admit that argument does have an expiration date, though.

Google bid ten million dollars for Spirit Airlines’ internal data. Mercor bid seven and a half, so there was a negotiation. The lot is roughly 100 million emails, about 500 million Teams messages, 7.5 billion passenger transaction records, and around 175,000 employee records that court filings date to 1986.

There’s no pricing model. There’s no vendor. The Greyhound bus of the skies is dead. A bankruptcy trustee has a duty to turn assets into cash, and it turns out employee email is an asset.

Every remedy I believe in requires a counterparty. Read the contract, ask the vendor, hold the renewal. None of that works against an estate. Nobody at Spirit signed anything covering this, because Google wasn’t party to anything when those people were hired. The record outlived the company. When the company died it went into the pile with the gate slots and the planes.

It hasn’t closed though.

A labor union went to court and objected that employee communications could be analyzed in ways that affect workers’ future job prospects, and that de-identification isn’t foolproof. The hearing moved to September 9.

That is the complete list of people who objected.

CFIUS, the DOJ rule and Mobley

KM — He’s conceding more than he thinks. If the buyer’s leverage disappears the moment the seller does, then it was never leverage. It was a courtesy that lasted as long as the vendor needed the renewal.

And the reason a union had to be the one to file is that the rule you’d expect to cover this says the opposite. I went looking, so I’ll save you the trip.

CFIUS doesn’t reach it. The regulation at 31 CFR 800.241 specifically excludes data maintained or collected by a US business concerning the employees of that business from the definition of sensitive personal data. Your workforce file is carved out of the foreign investment review by name. Every question I asked in the last section about whose money is inside your payroll system, the review that exists to answer it was written not to.

What does exist is the Justice Department’s bulk sensitive data rule, effective April 2025 with obligations from October. Six countries of concern. It bites at 100,000 US persons and reaches vendor agreements even when the data never leaves American soil, because granting offshore administrative access can be enough to trigger it. Below a hundred thousand people, outside those six countries, there is no general American requirement that an employer know where its employees’ Social Security numbers are processed or who owns the company processing them.

You can be lawfully ignorant. Most employers are.

The one place anybody is being made to answer is a courtroom. In Mobley v. Workday, Judge Rita Lin held in July 2024 that a vendor can be liable as an agent of the employer, writing that a third-party agent may be liable as an employer where the agent has been delegated functions traditionally exercised by an employer, and that Workday’s role is no less significant because it allegedly happens through artificial intelligence rather than a live human being. This June the court kept most of the discrimination claims alive and ordered Workday to produce a list of customers who had enabled the AI features.

That holding is about liability. Turn it around and it’s about architecture. If one vendor performs a function for thousands of employers, then one model with one threshold makes the same call everywhere it runs. The scale number is Workday’s own: in the certification order the court quotes the company’s filing saying 1.1 billion applications were rejected using Workday during the period at issue, and that any notice would still invite potentially hundreds of millions of potential plaintiffs. Workday volunteered that number while arguing the collective was too large to notify.

No court has held that being screened out by one algorithm screens you out everywhere it runs. I’m not going to claim it did. That’s my inference and I own it as mine. But look at the shape. Same model, same thresholds, 1.1 billion decisions, no notice, and no file anybody can request.

Subprocessor lists and change of control

LH — She didn’t move me on the sponsor question. Between a fund with a seven-year clock and a public board protecting a quarter, the answer depends on the fund and the board.

But I’ll double down that the software buyers who start with the cap table miss the pricing page. They should be reversing that approach, but still not ignoring ownership.

Mobley settles something else, though. It took Derek Mobley three years and a federal judge to establish a fact a buyer could have asked for in an RFP. Nobody reading this is going to file a lawsuit. So it isn’t litigation. And it isn’t a regulator, because Kristy just showed you the regulation with your workforce cut out of it.

That leaves the contract. It’s not much. Here’s what I’d look at.

Pull your subprocessor list. Find out who processes your employee data, by name, and when anybody last looked at it. Find your change of control clause and read what happens to your terms when the company sells. Then ask where your record goes when it does, and get it in writing.

Three questions. A CHRO can ask all three tomorrow.

KM — And the person whose record it is can’t ask a single one.

That’s the part Lance’s lever doesn’t reach. Everything he just described is a right that belongs to the buyer. The employee isn’t a party to the contract. They can’t read the subprocessor list, can’t invoke the change of control clause, can’t demand the retention answer, and won’t be told when the record moves. Spirit proved it. Forty years of email, one union filing, and not one of those 175,000 people had standing to say a word about it.

Ask your three questions. I mean it, ask them, because somebody with standing should. Just don’t confuse having the questions with having the right.

Fail all three and you don’t have a vendor. You have a landlord. Get all three answered and you still don’t own the building. You just know who does.

I was at our county fair last week, eating a deep fried Oreo, and determining which of my daughter’s friends looked sicker after riding the Scrambler. It was a good time and, for a brief moment, I wasn’t thinking about SaaS.

Then I opened my email and saw a note from Siobhan Fagan, the editor of Reworked. She was one of a few people who sent me a note when rumors of Workday going private started circulating. She was doing a story and wanted a quote.

I voice-to-texted my answer over screaming people and the drone of carnival sounds, cleaned it up, and sent it her way. The next morning when it was published, I put together a quick post while I was picking up my travel trailer for one last camping trip of the summer. I typed it up on my phone and dropped in two logos in Canva so I had a visual and called it a day.

And that’s the story behind my most popular LinkedIn post of 2026. A post about Workday, written on a phone on a Friday. It’s a day I rarely post anything because I’ve never really had much luck. 125k+ impressions? Sure, why not.

Shows you what I know.

But I’m not posting about the mysteries of the LinkedIn algorithm. The fact is, I’ve been thinking about the finances of big SaaS ERPs like Workday, Oracle, and SAP for a long time and I even wrote about it last month (that was, no joke, one of my least popular posts of the year). I thought the underlying financial strain was an under-reported story that was worth examining from the lens of a buyer or customer.

At least I occasionally get some things right.

I’ve been on record that I’m not on the SaaS-pocalypse train. It just doesn’t make sense to me to build some of this stuff yourself. That doesn’t mean things are going to be easy for these big ERPs.

The suggestion of such a massive enterprise going private (and the accompanying stock bump) demonstrates the state of large SaaS ERPs better than the thousands of words written about it. Everyone seems to be skeptical of an updated private equity playbook that doesn’t involve stripping off the parts that can be sold, loading it up with debt, and let some vulture take the scraps of the business when you’re done. I get it.

Thomas Otter takes a balanced position if you want to get into the nuts and bolts of why this one could (and should) be different. It’s a must read if you’re really into all that.

I don’t share the doomer position on going private that some of the people who have thoughtfully commented on my post do though. For one, even publicly-held companies are acting more like PEs than PEs. You just have to cross San Francisco Bay from Pleasanton to find Oracle, which is up to its eyeballs in debt and will be laying off thousands more to fund more AI.

Even though Treasury Secretary Scott Bessent said the overall economy isn’t K-shaped anymore, I believe companies like Workday, Oracle, and SAP are still in a real fight for which leg of the K they will be following. We’re still early but we have some clues about what doors are still open.

If I were immersed in the Workday ecosystem, I’d be less concerned about a take private deal that I don’t know the details of. I’d be much more worried about what happens if a deal doesn’t happen, Workday stays public, and they have to figure out how to win the right way. They can’t outspend Oracle, which is funding $40B of their infrastructure spend with debt and equity. While the market had been a little more forgiving with Oracle’s free spending ways earlier, their stock is now down 25% YTD.

A PE partner with a plan, resource alignment, and good leadership structure without the pressures of Wall Street gamesmanship sounds like as good of a bet as what we know about Workday’s public plans. Another wrinkle? We’re less than two months away from Workday Rising, which will put thousands of customers in front of the company where questions will be a lot harder to ignore.

As someone without a dog in the fight, it will be interesting to watch. The major ERP and HCM players are all hitting headwinds, though it is affecting them in different ways. What will this category look like in five years? I wish I could tell you (and make the appropriate stock purchases).

It might not beat a deep fried Oreo, but it’s an incredibly important story for every company you know the name of.

What else is happening this week?

• AI’s Effect on Workplace Culture. In workplaces that rolled out AI, 25% of employees say the culture got worse against 24% who say it improved. Morgan Meinen and Megan Mulherin find the split is the manager: 33% versus 4% on whether work actually changed.

• Is Your Culture Ready for AI? The Question Most Organisations Haven’t Asked. Mervyn Dinnen and Nick Holmes describe an AI-ready culture as one built on trust and purpose rather than an efficiency pitch. The bill for skipping that: 29% of employees say they actively undermine their employer’s AI strategy, 44% among Gen Z.

• AI Hiring Tools Can Invent Their Own Bias, Research Finds. Princeton and Chicago gave four fictional groups identical odds and the models still sorted them into different jobs, per Jill Barth. Only an explicit diversity instruction fixed it.

• What Happens When the Disruptor Becomes the Incumbent? Let’s Talk About Workday. Steve Smith‘s read on the Silver Lake rumor: investors don’t flash $50 billion at dying companies, they flash it at undervalued ones. It’s a good take on top of his usual good newsletter fodder.

• Steal My Claude Setup. Laurie Ruettimann published the instructions she gives Claude so it edits like a partner instead of a ghostwriter. About 72 people marked up her last book, me included, and none of us got a watermark.

• The Job Board Is Dead, Long Live the Job Board. Applications per posting doubled to 244 since 2022 while recruiting teams shrank by half. The fix Virginia Backaitis finds everyone arriving at is referrals: 7% of the pool, half the hires, older than job boards.

• The Standard We’re Setting for AI Should Apply to Humans Too. Master Burnett takes Madeline Laurano‘s 41% who want full explainability from an AI interviewer and asks whether they could explain last week’s human decision. Only 17% wanted validation against job performance.

• AI Governance: 1 Intro. Boards sign off on AI without anything close to the reporting they demand from accounting. John Sumser opens a new series with the evidence package that gap calls for, down to overrides and near misses.

• Leaving Work on Time Is Not a Character Flaw. Madison Butler did the math on the extra 10 or 15 minutes at the end of the day: about 1,625 hours over 25 years. Her line is that your employer bought your labor, not your life.

• What We’re Losing by Using AI for Everything. Mary Faulkner leaves the Zuckerberg demolition to The Verge and keeps the harder question for herself. There’s always a trade between efficiency and meaning, and somebody has to say out loud whether it’s worth taking.

• Payscale Survey: Employers Plan 3.5% Raises for 2027. Inflation hit 4.2% this spring, so a 3.5% raise is a pay cut. David Manaster notes that a quarter of the 1,266 employers surveyed already believe weak raises cost them people, with another 22% unsure either way.

• Extinction Has a Funny Way of Looking Like a Hiring Spree. SHRM’s CEO stood in front of humanoid robots and told a room of HR professionals their field faces extinction. Daniel Space calls that the language of a funeral describing a transformation.

• Despite Massive AI Investment, Jobs in HR Are Booming. The Lightcast numbers behind that pushback are Josh Bersin‘s: HR postings up 6% over the last 24 months. He reads it as administrative HR work turning into consulting work, not disappearing.

• The Edge of Now Episode 3: AI Can Find the Answer. Can Your Organization Still Think? Jess Von Bank and Jason Averbook want organizations to quit leading with prompt engineering. Their measure is whether you’re making better decisions, not whether you’re faster.

• How Does Your Demographic Background Influence What AI Tells You? The study Steve Hunt surfaced found models tuning answers to what they’d worked out about the user, down to gender, income and location. Salary advice came back lower for women and for people of color.

• The State of Workforce Compliance and Employee Trust. Compliance lands in the employee experience whether or not anyone planned it that way, which is the thread Steve Boese pulls with Experian’s employer services side.

• AI Makes Intelligence Cheap. Judgment Becomes More Valuable. AI in recruiting is turning into infrastructure, closer to electricity than another app in the stack. Kevin Wheeler wants recruiters knowing when to question an output, not how the model was built.

• AI-Generated Content Is Entering Its Google SEO Moment. LinkedIn now lets users flag low-value AI posts. Nicholas Biron sees that as the platforms starting to correct, the same way Google eventually came for SEO gaming.

• When I Choose Work. At 17 the choice was a graduation dinner or a shift at American Eagle. The shift won, and twenty years on Kat Kibben concludes work never chose them back.

• ‘Did They Ghost You?’ Website Takes Aim at Employers Who Go Radio Silent on Candidates. More than 2,000 companies are listed, Apple and Microsoft among them, and at peak it was taking six reports a minute. The guy who built it told Paige McGlauflin that even a terrible final interview earns you a reply.

• 4 New Legal Questions Being Raised by Background and Identity Checks. Whether AI-assembled candidate profiles count as consumer reports under the FCRA is one of four fights Bob Quackenboss and Evangeline Paschal map out. Disparate impact is the one heading for the Supreme Court.

• Average S&P 500 CEO Pay Surges 21% to Record High as Musk’s Mega-Deal Sets a New Standard. Average pay climbed 21% to $22.8 million with Musk excluded, against mean worker wages of $69,770. That’s 312 to 1, or 5,387 to 1 once you put Musk back in.

• Google Insiders Warn Job Candidates About Its AI Hiring Filters. A team inside DeepMind told internal candidates to file a second form, because there’s a non-trivial chance the system screens them out by mistake. Jen Colletta on posting a rough road sign instead of filling the potholes.

• Bank of America Won’t Let Employees Work Remotely 2 Days in a Row. Gabrielle Saulsbery reports the new rule closes the Friday-plus-Monday gap for staff who already owe three days in the office. Pick a lane: hybrid with one simple rule, or everyone in five days.

• The Last Mile of Skills Transformation Is the Manager. Quarterly career conversations went from 8% in 2019 to 63% today in new Fuel50 research from Anne Fulton. Managers need that data in the conversation, not a quarter later.

• What We Leave Behind. The 8:15 meeting exists because an executive who retired in 2021 had a commute, five years after he left. Robin Schooling writes about that kind of residue, and how little of it any knowledge management system will hold.

• HC Insider: August 2026. Seat licenses made adoption the metric worth chasing. Tokens make it impact, which is the reframe Kyle Forrest leads with this month.

• Sourcing Faster Without Losing the Human Part of Hiring. 93% of hiring managers say they’ve lost a strong candidate to a slow process. Dan Schawbel works that gap with Indeed and a solo recruiter, on a tool saving about seven hours a week.

• How HR Can Own Workforce Transformation in the AI Era. Five podcast episodes condensed into one read by David Green. The neuroscience one has teeth: resistance is a protective response with six diagnosable causes.

Have a great rest of your week!

I didn’t publish last week. Did you miss me?

One of my worries when I preload posts and newsletters before I am offline is that something bad will happen and my dumb automated posts keep coming through whatever regrettable tragedy we are dealing with. So one of the things I decided was that if I was truly offline for a break (like I was last week), I wouldn’t automate anything.

I appreciate the ability to truly disconnect. That means I have some catching up to do, which is always fun.

One of the things I thought about while I was trying to avoid wildfire smoke in the PNW is how squishy the difference is between good human work that’s AI assisted and good AI work that’s human assisted. How much are humans really in the loop? How much are people understanding the work that they are producing? Does it even matter?

That last one is more of an existential question.

I filed my column for Reworked on this topic before I left but I still had it stuck with me as I was on vacation. And it is more than separating human and machine.

The research from KMPG and UT Austin found that while nearly three-quarters of folks pushed back on AI supplied outputs (as you would hope), a third of those folks were completely ineffective in their pushback and actually performed worse than the people who copied and pasted Claude’s answer verbatim and moved on.

Managers have a hard time telling the difference too and, maybe more importantly, guiding people to use AI in a productive way. For as long as we have trained managers, we’ve focused on output quality as the North star. Do good work? Get a pat on the head. Good boy, now sit. Do bad work? Then we will talk about what you did wrong to get there and how to solve it.

Now that AI is obfuscating bad work behind the veneer of plausibly decent work, you can’t just judge on output alone because it is really tough to figure out. It was much easier when you could count em dashes or look for certain constructions to understand tells.

Those times of easy detection are disappearing (although Anthropic is trying to layer in some of that transparency).

What can leaders do to figure out if people are actually understanding and doing the important parts of the work? I know this is a radical idea but let me have this.

You talk to them about the work. Wild, I know.

You can read the rest of my ideas over on Reworked.

What else is happening this week(ish)?

• North Korean IT Workers Use Real-Time Deepfakes to Beat Hiring Checks, Eleven Nations Warn. Eleven countries just warned employers that North Korean operatives are running live AI deepfakes through job interviews to land remote IT jobs, then stealing source code once they’re hired. This stopped being a niche threat a while ago.

• ADP Leader: Most HR Leaders Are Asking the Wrong AI Questions. Joe DeSilva at ADP has a piece of AI advice actually worth following: stop asking where it can save time and start asking where it creates real business value.

• While Layoffs Slow Overall, Job Losses in Tech Soar. Layoffs overall looking better isn’t the same thing as layoffs being over: total cuts are down 46% year over year, but tech cuts jumped 67% in July alone.

• Who Controls the Past Controls the Future. The executives blaming AI for this round of layoffs are the same ones who said the pandemic hiring surge would last forever, and Laurie Ruettimann isn’t in a hurry to trust their timing twice.

• Consequence Architecture. Chris Havrilla wants companies to get more specific than “human in the loop,” naming one actual person who owns every AI agent’s outcome before that agent runs. Worth doing: 88% of companies use AI somewhere in the business, and only 8% have anything resembling real governance over it.

• Breaking the Applicant Funnel. The applicant funnel has been broken since AI let candidates flood it with volume, and Steve Hunt wants to replace it with something closer to a tournament, where candidates know the rules and get real feedback.

• Economists Expect a Cooled Labor Market and an AI Reshuffling of White-Collar Work. Laura Ullrich and Svenja Gudell just launched Indeed Hiring Lab’s new quarterly survey of more than 100 economists, and slightly more than half of them expect AI to be a net drag on jobs over the next year.

• We Made AI Compulsory. Now We’re Making It Contemptible.. Every platform built “write with AI” into the toolbar, and now those same platforms are shipping detection features to shame you for using it. Jess Von Bank‘s line for it is right: human error is becoming a CAPTCHA.

• Trump Administration Expands Paid Leave Tax Credit for Employers. Treasury expanded a paid leave tax credit in a way that actually makes sense: employers can now claim it against insurance premiums instead of just wages paid directly, which should make it a lot easier for small and mid sized companies to use.

• Psychological Safety at Work May Depend on the State of DEI Outside of Work. Barjinder Singh‘s new study found that workers who see their community as inclusive report feeling psychologically safer at work, even at companies that aren’t doing much to earn it. Your zip code is apparently doing some of your employer’s job for them.

• Strategic HR Leader: A Title Everyone Claims, But Many Can’t Define. Every HR leader wants to claim the “strategic” title, and Jennifer McClure‘s point is that getting invited to the executive meeting was never actually it. Her data backs it up: AI meaningfully boosts HR’s strategic impact at some companies and does almost nothing at others using the identical tools.

• Starbucks Is Ending GLP-1 Coverage for Weight Loss as Employer Costs Climb. Sarah E. Needleman reports Starbucks is cutting GLP-1 coverage for weight loss starting in October, now that those drugs eat up 11.4% of employer health claims, up from 6.9% in 2023. Expect more employers to follow until the price comes down instead of the coverage.

• Slop Filters, Startup Millions, and AI-Powered Pink Slips. The Chad & Cheese roundup has a stat that matters: 43% of managers at large companies now let AI make layoff decisions with nobody checking its work.

• Google’s Answer to AI Uncertainty: 15 Million Real Conversations, Mapped. Google mapped 15 million AI conversations across more than 800 occupations and found the technology touches 88% of US employment but does work in only about a fifth of the average job’s tasks. Jill Barth‘s rundown of the report backs the boring truth over the loud one: adoption is broad, but it’s shallow.

• AI Is Driving Down the Cost of Outplacement. Chieh Huang‘s pitch for his AI outplacement startup Pelgo is blunt: no coffee shop on earth was ever buying its laid off baristas a $2,000 outplacement package, but at $20 a month, some of them might. Whether cheap outplacement actually helps anyone land a job in this market is still an open question.

• CFOs and CIOs Starting to Treat Enterprise AI Like Traditional Tech: Is It Cost-Effective, Does It Work, and What Are the Risks?. CFOs were never going to stay sidelined on AI spending forever, and Josh Bersin says the hype cycle is ending in the most boring way possible: they’re demanding the same cost justification they’d want for any other software.

• You Can’t Tell If You’re Learning Anymore. Jason Averbook‘s argument is that producing the work still matters even when AI can do it for you, because the struggle of doing it yourself is what used to prove you were actually learning something.

• McKinsey Is Wrong. The Great Flattening Is Mostly a Relabelling Exercise. Toby Culshaw looked at the org chart everyone’s passing around, the one that flattens two management layers into oblivion, and noticed it mostly just renames Directors as Team Leads. Gallup’s numbers back him up: average span of control went from 8.2 direct reports in 2013 to 12.1 now.

• New Podcast from John Vlastelica. Congrats to John Vlastelica on finally launching a podcast, which honestly should have happened years ago given how good he is talking about, well, anything.

• A Sonnet in Your Inbox. Robin Schooling breaks down the farewell email as a stricter structure than most actual writing anyone does at work: an announcement, a gratitude section that doubles as commentary on who mattered, a highlight reel, and a deliberately vague sentence about what’s next.

• The People & Progress Report: Let’s Flood the Zone. Researchers ran identical salary negotiation prompts through five AI models, changing only the persona, and a male candidate got coached to ask for $400,000 while the identical female candidate got coached to ask for $280,000. Kristy McCann‘s newsletter has nine more like that this week, plus a genuinely good summer playlist.

• The How #29. Katie Achille‘s metaphor for leadership since becoming CEO of her PR firm this spring: some leaders are sails, providing agility, and some are anchors, providing stability, and the hard part is knowing which one a moment needs. I like that.

• Plausible Deniability With a Subscription Fee. The EU delayed its high-risk AI employment rules again, pushing accountability requirements to December 2027, and Jackye Clayton has the line that captures what that delay actually protects: “We built a hiring process that never required anyone to look.”

• Work Will Always Be There. These Moments Won’t.. Your company will cut you the moment AI makes it cheaper, and Mike Wood‘s advice is to do great work, then close the laptop and go be with the people who actually know your kids’ names.

Have a great rest of the week!

“The employer-employee relationship is broken.”

In a trip through my inbox over the past 15 years, I have seen countless pitches that told me exactly this. Employees and employers have no sense of collective loyalty anymore. Engagement continues to tank. Longevity at companies is a rarity. Hiring is tougher than just finding the skill set you need.

After seeing coworkers get laid off with no notice, little severance, and thrown out into a brutal job landscape, it’s understandable that people might think to just say no to a new job in general. “I’ll just contract for awhile.”

After seeing employees leave with little notice and switch allegiances to a competitor, it’s understandable to that organizations might think twice about employing another person who can just leave. “We’ll just contract this role for awhile.”

I’ve been told for a long time that this shift is inevitable. That AI might accelerate it. That it might be healthier for our psyche if we disconnected our identity from our work.

And I nodded along. It makes sense in a sort of “no shit” way.

Employment is risky for employers. There are a lot of rules. There are liabilities and taxes. And even if you look past that, even a less ideal hire can really hurt an organization.

And employment isn’t always that great for employees either. Getting a job is usually a pain. With minimal protections for U.S. workers compared to much of the first-world, it can feel like contracting is just taking out the middle man.

But after nearly half a year of solopreneurship, I don’t think that argument hits as neatly as before. Which is a weird way to say that after leaving a steady paycheck, I believe the value of a good employer-employee relationship is severely underrated by too many people on both the employer and employee side of the equation.

I’m not playing Devil’s advocate here. I wrote a whole researched piece about it in Reworked last week.

Having a job is valuable. Having employees is valuable. In a good relationship, both parties do better together than they do separately. I’d say that in most instances, having a job is better for most people and that having employees is better than just simply relying on gig workers.

The why comes down to continuity and judgment. Employees anticipate and judge better when they’re at a job for a longer time. You know how things run. You know expectations. This is good for us and our mental wellbeing.

And those things are really good for employers too, even if they don’t always recognize it.

That doesn’t mean that jobs are perfect or that our reaction to disassociate from them because they are hot piles of garbage isn’t a natural response. So I also shared some ideas that would make them better, like better protections for workers and rethinking management practices.

But there were two things I didn’t write about.

One is that it’s really easy to take a good job and a good employee for granted. That’s not a flaw in the system but proof that a good employment relationship can be so good, it just feels like it will never end.

Two is that regardless of how you feel about employment, we have to do something to fix running solo if that’s what you choose. Taxes and regulations for the self-employed are onerous. Benefits are ridiculous. The lack of stability and certainty can wear on you.

In any case, give it a read and let me know if you agree or disagree.

I’m on Up Next @ Work

A few weeks ago, I recorded a podcast with HR tech legend Jeanne Achille to talk about the current version of the weird world of work and technology. We also talk about jobs might look like in a few years when my kid is old enough to decide what she wants to do.

Also, a note: If you want to bring me onto your podcast, just promise me we can talk about basketball for about two minutes and I’ll be good. You can even cut it from the show.

Take a listen.

What else is happening this week?

• US Jobless Claims Tumble to a Decades-Low 187,000 as Layoffs Stay Historically Low. Claims haven’t been this low since 1969, which reads great until you notice employers added just 57,000 jobs in June and the unemployment rate fell mostly because people stopped looking for work. A job market nobody is leaving and nobody is entering isn’t the same thing as a healthy one.

• Stop Managing Machines Like People. Stop Managing People Like Machines. Chris Havrilla launches a new Substack and a new venture with the argument that we spent thirty years measuring human output instead of outcomes and are about to do the exact same thing to agents.

• 50 HR Tech Demos and Some Tough Love for Founders. After sitting through roughly 50 demos, Anita Lettink tells founders that the problem that annoyed them at their old job is probably not a strategic priority for anyone else. Her warning about chasing enterprise logos before you have revenue is a good one too.

• The State of Candidate Fraud Detection & Prevention. A North Korean operative passed every video interview at KnowBe4, a company that sells deception training, and got caught by endpoint security on day one. The rest of the data on where fraud defenses cluster versus where fraud actually succeeds comes from Kyle Lagunas.

• Resume Fraud Is Real. John Sumser points out that many of the fraud numbers everyone repeats lump title embellishment in with deepfake interviews, and that the vendors publishing those numbers often sell fraud detection. So, yes, it exists but buyer beware.

• The Best HR & People Analytics Articles of July 2026. BCG’s finding that the capabilities executives value most are exactly the ones AI erodes fastest would be worth the click on its own. It is one of about fifteen things worth reading in the July roundup from David Green.

• What You Choose to Interrupt. Seven pieces in, Jess Von Bank ends the Human Thesis series on a clear point: most companies are automating dysfunction they never bothered to fix. Capable technology in a broken operating model just gets you to the wrong answer faster.

• He’d Already Lost Before He Started: What Zuckerberg Got Wrong on AI Agents. Stacey Harris reads Meta’s stalled agent progress as an org design problem, because it ran as an IT project and skipped the context and auditing work HR would have insisted on. Her data backs it up: only 4% of organizations are running full HR agents, and the early movers mostly sped up bad processes.

• Closing the AI Hype Gap. The most AI-exposed workers saw a smaller unemployment increase than the least exposed between 2022 and 2025, and Goldman puts AI-related job losses at 2.5% of US employees. Gary Hamel and Michele Zanini line the evidence up against the three biggest claims, and it certainly beats the hype.

• Only 30% of Americans Are Working in the Middle of the Workday. More than half of Americans are on leisure or sports by 8pm, and the middle of the workday is a lot emptier than any return-to-office policy assumes. The American Time Use Survey chart shared by Todd Nicolini deserves a long look.

• Netflix Executive Says Sharing Mental Health Struggles at a Leadership Retreat Led to His Firing. A Netflix VP says he disclosed physician-supervised ketamine therapy for depression during a “Vulnerability-Trust” exercise at a company retreat, got investigated two months later, and was fired in April. These exercises were always cringe, and now there is a lawsuit testing whether anything you say in one is protected at all.

• Trump’s Rollback of DEI Is Transforming Who Runs America’s Biggest Companies, Study Finds. Women and racial minorities made up 40% of new S&P 500 director appointments over the past year, down from 72% in 2021 and the lowest share since 2014. Boards are backfilling with sitting and former CEOs, a pool that is less diverse by definition, so the 49.3% of seats currently held is a lagging number.

• The Front End of Hiring Is Where Adjacent Talent Is Won or Lost. Master Burnett argues in part two of his talent discovery series that AI can’t find adjacent talent because most organizations never defined the work in the first place, they just recycled a job description. A better model will not fix a bad intake conversation.

• What Oregon Winery Leaders Taught an I-O Psychologist About Complexity. Agriculture, heavy regulation, public-facing branding, and multi-year paths to value all sit inside one operating problem in the wine business. Gabby Burlacu moderated the panel, and the takeaway is that tech is not nearly as complicated as it thinks it is.

• All In Before the Flop. OpenAI and Anthropic now have to IPO near a trillion dollars or read as failures, which is a strange bar for companies still proving out their business models. Everyone pushed their chips in before seeing the flop, and as Andrew Hilger notes, it’s usually the player who switched to water hours ago who takes the pot.

• Everyone Is Back in School Now. Jason Averbook points at the PNAS study where students using ChatGPT scored 48% better with it and 17% worse without it afterward, then asks what happens when you cut the analyst program to pay for the tools. Productivity borrowed from next year’s capability isn’t real productivity.

• EEOC Gets One Step Closer to Eliminating Equal Employment Opportunity Data. The commission voted to repeal 60 years of EEO-1 demographic reporting on the theory that collecting the data without a discrimination complaint attached is itself the problem. Employers still have to collect it under Title VII, they just lose the early warning that told them where their exposure was.

• When a Founder’s Opportunity Fund Win Ends in a 50-Page RICO Suit. Win the competition, get summoned to an eight-hour twelve-on-one interrogation, watch the valuation get cut, then get threatened on Christmas Eve. Kristy McCann walks through the whole loan-to-own playbook on the Chad & Cheese Podcast and somehow makes it oddly funny.

• Talent vs. Tokens? It’s About Bench, Not Cost. Junior analysts were never hired for the deliverable, they were hired for the reps, and the zero-to-70% stretch where those reps live is the first thing an agent takes. That makes it a work design problem before it’s an AI problem, which is the whole argument from Steve Smith.

• You’re Not Broken. The System Is. Twenty years of saying the things HR won’t say, and Laurie Ruettimann is now wondering out loud whether any of it worked. Meg Bear and Amy Wilson get her to the good stuff, including the case that financial freedom is what actually buys you the room to be honest at work.

• America’s Layoff Safety Net Was Built for a Different Economy, and AI Is About to Break It. Bijal Shah makes the case that unemployment insurance, outplacement, and WIOA were all designed for downturns that eventually end. Only 18 states had adequate UI reserves in January 2025, down from 31 before the pandemic. That should concern anyone.

• How “No Tax on Tips” Is Starting to Disrupt Frontline-First Companies and Workers. The tips exemption is now a $4.5 billion transfer that shows up, by Josh Bersin‘s math, as roughly 35% lower labor costs for companies that collect tips than for the ones that just pay wages. Target and Costco didn’t do anything wrong here, they just don’t have a tip line.

• Segal Survey Projects Health Plan Cost Trend to Reach 15-Year Historic Highs. Medical trend is projected at 9.9% and prescription drugs at 11.5%, the highest in 15 years, driven by specialty drugs and GLP-1s. About a fifth of inpatient cost growth is attributed to AI-assisted coding intensity, which tells you what that technology is actually being used for.

• Ford’s Rehire Wave Has HR Leaders Rethinking How AI Layoffs Get Decided. Ford rehired 350 veteran engineers after automated inspection missed defects experienced staff would have caught, and roughly three in ten employers have now run some version of that round trip.

Have a great rest of your week!

One of my persistent annoyances in my never-ending quest to understand what is going on in the world of work is how fragmented all of the publicly available information is to access. We have some seriously great data resources and they are all available in different ways on 15,000 different websites.

Okay, that’s a slight exaggeration.

One of the first things I built with Claude Code was a way to access all of this information in a single view. With API access for nearly all of the data, it was pretty easy. I got a little dashboard that lived on my computer that I could browse and search in one place.

Then I wanted to chat with my data. First, I added in a chat into the dashboard. As I learned about MCP servers, I thought it would be better to just be able to chat with the data. So I built that too.

Then I simply used it. It has helped me grab numbers for research, columns, and reporting.

I told a few people about it and they encouraged me to share it. So, with the help of Claude’s Fable, I improved the portability and interface some while adding a few latent features that I was just using workarounds for.

So, today I am launching my first research tool: Workforce Data Explorer (killer name, right?).

What is it and how can you use it?

Workforce Data Explorer is an open source tool for understanding the U.S. labor market.

U.S. labor data lives across a dozen agencies and sites, each with its own interface, format, and limitations. This puts it behind one interface:

• 89 curated sources, searchable by topic, geography, and update frequency

• 30 data sources connected to live APIs - the tool fetches current numbers on demand, not links to somewhere else

• Two kinds of data, deliberately mixed:

  • Official statistics

  • Faster signals from other open sources

There are three ways you can get access to it:

1. The dashboard — zero setup

You can access a live dashboard here: workforce-data-explorer.streamlit.app. This is the easiest way to see what it actually looks like and how it works.

It opens on a live overview: unemployment rate, monthly payroll change, job openings, and weekly UI claims, each with a two-year sparkline. From there, ten pages:

• FRED Time Series — a front end to the Fed’s 816,000-series database: 70+ curated labor topics (quits rate, wage growth, participation by demographic) plus free-text search of the whole thing

• BLS Series — jobs-report series, JOLTS, Employment Cost Index, productivity; enter any SOC code and get that occupation’s mean wages and employment

• State Labor Markets — pick a state, get its four headline indicators on one screen: unemployment, payrolls, weekly claims, JOLTS openings and quits

• Census — three tabs:

  • ACS demographics (employment, income, commuting, remote work) by state or county

  • Quarterly Workforce Indicators — hires, separations, earnings over time

  • County Business Patterns — employment by industry sector

• O*NET Occupations — profiles of 900+ occupations: skills, daily tasks, abilities, technology used — plus an “AI usage by occupation” view showing how much of each job’s real task list appears in actual AI usage. Measured behavior, not predictions.

• DOL Enforcement — weekly UI claims, OSHA inspections searchable by state and date, wage-and-hour cases with back wages assessed

• SEC Filings — full-text EDGAR search: layoff 8-Ks with company names and links to the filings, human-capital disclosures from 10-Ks, any company’s filings by name

• Job Postings — Indeed Hiring Lab’s daily indexes, the timeliest public read on labor demand that exists:

  • National trend, all 50 states ranked, 41 occupational sectors ranked

  • Any metro area over 500k population

  • The share of postings mentioning AI, tracked over time

  • Everything indexed to Feb 2020, so every number is a plain “vs. pre-pandemic” comparison

• Catalog — the searchable index of all 89 sources

• AI Assistant — ask in plain English (”show me quits vs. openings since 2022”), it fetches and charts the data. Rate-limited on the shared instance; unlimited with your own free key though it is slower than your typical chatbot.

2. Plug it into your existing AI

If you want to chat with the data and you already have an LLM that has MCP capabilities, this is the best way.

Setup is one URL (updated!): https://workforcemcp.beaconturn.com/mcp

• Claude: Settings → Connectors → paste the URL (paid plans)

• ChatGPT: Settings → Connectors, developer mode → same URL

Your AI gains 15 tools it calls live while answering you:

• FRED and BLS time series, with search

• Job postings rankings by state, sector, or metro

• AI share of postings

• Full state labor-market snapshots

• Occupation wages and O*NET profiles

• AI-usage exposure for any job title

• Weekly UI claims

• Layoff filings and company filings from EDGAR

• The source catalog itself

Ask is a workforce data question and it pulls the actual claims and postings numbers from live data that is accessed via a very efficient API poll rather than letting your LLM roam freely on the internet.

3. Run it yourself

Clone the repo on Github and the whole stack is yours:

• The dashboard, the MCP server, and the data layer underneath

• Grab your own free API keys, ~5 minutes of signups, all listed in the README

• Deploy configs included for hosting your own copies of both the dashboard and the connector

• A one-click desktop extension (.mcpb) that connects the local MCP server to Claude Desktop with no config-file editing

• The connectors are plain Python modules with no framework lock-in

The whole thing is MIT Licensed, which means you can fork it and generally do you what you want with as long as the license text is included in your work.

Find a something wrong? Open an issue on GitHub, with quick templates for bugs, data problems, and source suggestions. Or you can hit me up if you think I am missing anything else that should be included.

What else is happening this week?

• Your AI Is Hiring Too Many Bad Employees. The real argument isn’t about tokens, it’s that companies stuff AI prompts with the same junk they used to stuff into meetings. Cut both, says Jason Averbook.

• The How #28 - July 16, 2026. Katie Achille watched someone she knows get swallowed by a viral news cycle and turned it into a case for treating your media diet like mutual aid. Stop feeding stories that aren’t yours to feed.

• The $21 Billion Company Behind Eventbrite, Vimeo, and AOL Has Just Revealed How It Picked 286 Out of 800,000 Job Applications. Bending Spoons hired 0.04% of 800,000 applicants last year, and treats it like a brag, not a crisis. They’re right that interviews barely predict who’s good at the job, though.

• Meta’s AI-Based Layoffs Allegedly Targeted Workers Who Had Taken Protected Leave. Twenty-six Meta employees say the layoff algorithm scored them on metrics you can’t rack up while on medical or pregnancy leave. Meta says people made the calls, not AI, an odd claim about a system built for exactly that.

• The U.S. Labor Market May Soon Face a New Crisis: Too Few Workers. Pick a panic: AI eliminates half the jobs, or boomers retire faster than anyone replaces them and employers spend the 2030s begging for workers. Door two comes with real wage leverage for the young workers who stick around, according to demographer Steven Ruggles.

• Starbucks Says Coffee Is For Vibe Coders. Steve Smith flags the true story in Starbucks’ $400 million software budget: it’s vibe-coding its own inventory and maintenance systems instead of paying Microsoft and IBM. When your biggest customers can just build it themselves, every enterprise vendor’s pricing model gets shakier.

• There Are No Dumb Questions (But There Are Some Very Expensive Ones). The question that nearly tanked Jason Seiden‘s first job wasn’t dumb, just too honest: he asked his CEO to prioritize his workload and got told to figure it out himself. Every question you ask reveals how much you understand about the room you’re in.

• Indeed Is Walking a Fine Line on the Fine Print. Indeed’s new terms insist it’s not, and never will be, anyone’s legal agent, a move aimed straight at the Workday and Eightfold discrimination suits. Two bullet points later it launched a bot that applies to jobs for you and bills the employer per application, a contradiction Mike Wood catches immediately.

• The Middle Miles: What Nobody Tells You About Month Seven. Brian Fink is seven months into grieving his father and argues American bereavement policy gets the shape of grief backwards. Three days of PTO assumes the danger is the raw first weeks, when it’s actually the smooth, forgettable middle.

• Study Finds Fair Workweek Laws Improve Schedule Predictability. Requiring two weeks’ notice and banning clopening shifts actually gives retail workers more predictable schedules, and Harvard’s Daniel Schneider found no evidence employers clawed the cost back in lower wages. Enforcement is incredibly important: New York City’s aggressive follow-through got a 25-point jump, Philadelphia’s got five.

• Workforce Atlases. Eleven states already have more retirees than kids waiting to replace them, and Utah is the only one still shaped like an actual population pyramid. It’s a genuinely useful map of where AI’s impact lands hardest, built by John Sumser.

• Humanity in AI Awards. Nominations close July 24 for MeBeBot’s award honoring HR and IT leaders who deploy AI without treating employees as collateral damage. Three winners get $500 and a badge, a modest prize for what should be the baseline.

• Deel Moves to Strike ‘Spy’ Witness in Rippling Lawsuit. Deel wants a judge to toss the core claims in Rippling’s lawsuit because the witness who says Deel paid him to spy won’t testify anymore. Rippling calls it a Hail Mary, and this is the rare corporate suit that deserves to actually go to trial for entertainment purposes.

• Celebrating 75 Years of Shaker. Joe Shaker Jr marked 75 years of his family’s recruitment marketing company, a rare milestone in an industry this consolidated. Congrats to the whole team on staying independent this long.

• HC Insider: July 2026. This new monthly Deloitte roundup, from Kyle Forrest, opens with a stat: 84% of organizations grew AI access without redesigning a single job around it. Adoption and adaptation aren’t the same thing.

• ERE Recruiting Innovation Summit Agenda. The November lineup is stacked with talent acquisition leaders from Ford, Wells Fargo, CVS Health, and Expedia. Nice job David Manaster and team.

• Energage and Engagedly Join Forces to Accelerate the Future of Talent Management and Workplace Experience. Two of the most confusable names in HR tech just solved the problem by becoming one company. Now there’s only one name to get wrong.

• BI Worldwide Buys WorkTango to Reach Mid-Market With AI Survey Tools. WorkTango keeps its name and team, but the real prize for BI Worldwide is a mid-market foothold it would’ve taken years to build, as board chair Gary Hansen made explicit.

• Why We’re Building the Harness, Not Another Agent. Rob Catalano‘s next act after WorkTango bets GTM teams need what developers already got: not a smarter agent, but the scaffolding underneath it. Gartner expects 40% of agentic AI projects to get cancelled by 2027, and his argument is that this is exactly why.

Have a great rest of the week!

Unlike some folks I’ve had the misfortune of listening to or reading, I don’t like saying that people chose HR because of an aversion to numbers.

That’s certainly not why I entered HR way back in the day. My math scores and aptitude were much higher in school than in any other subject, including writing.

But like my current world of research, media, and marketing, I found the work of HR to be much more interesting than the incredibly important work of finance and accounting. The applied math is still there and still challenging, but it’s not everything.

If you are a leader in HR, you can understand the basics of finances and while you may focus on your own org’s cash money, you might be ignoring something can really hurt your capabilities: the financial position of your HR tech vendor.

I know a lot of people read George LaRocque’s coverage of market movements and see it as inside baseball. I couldn’t disagree more strongly. And I think if HR leaders thought a little bit about how incentives, trust, and stability work at their vendors, they would understand the stakes better.

That was my argument in Reworked last week. That a company’s financial position should be a formal part of your purchase evaluation for a new (or renewed) piece of tech in your stack.

Why? Stick with me and you’ll see this is less about numbers and more about people.

Enterprise software as a whole is struggling. It now trades at a discount compared to the S&P 500. That never happens.

Multi-quarter stock declines are not great for investors but they are REALLY not good for a stock grants as a retention tool strategy. Worthless stocks make it easier for top talent to walk away, especially if they don’t see the stock rebounding anytime soon.

But it’s not just about retention. Companies with long-term stock dips do stupid things to try to reverse course. They may make a hasty layoff or acquisition, hoping to buoy earnings or Wall Street perceptions. A survey found that 78% of financial executives would sacrifice long-term value to smooth earnings and 55% would skip a positive, longer-term project instead of missing a quarter.

For customers, neither one of those outcomes points to a positive experience with critical software. While we can pretend like it shouldn’t matter, that willing a good product and customer experience can happen in spite of those pressures, it is harder when that hope meets the hard number reality.

While my article focused on stock performance, private, equity- or venture-backed organizations have similar risks. If a startup gets acquired by a larger company, will the product survive? If a company is struggling to attain series B financing, do people start jumping ship? If your favorite company goes private or gets rolled up into a private equity grouping, will you see service or product development decline?

While talking about RFPs and financials is a sure way to probably tank readership this week, I do think it is that important. And with all of the volatility in the market, it’s good to pay attention to it. Especially if you’re signing a contract that might outlive the team you’re working with today.

What else is happening this week?

• Peak Workforce. The American labor force peaked in November 2025 and has been shrinking since, the first drop outside a recession in the modern record. John Sumser is one of the only people running the actual numbers instead of just reacting to headlines.

• The Biggest Workforce Shortage in U.S. History Is Being Overshadowed by AI. Lightcast projects a six million worker shortage by 2032, and Jill Barth argues the AI panic is distracting HR from the bigger problem: there won’t be enough people left to hire.

• The View from Inside the Room. Jess Von Bank asked senior HR leaders anonymously how ready they felt to lead AI transformation, and the average score came back 3.4 out of 5. That’s a more honest number than you’d ever get in a town hall, and it should worry you more than another adoption chart.

• Quality of Hire vs Quality of Candidate. Most hiring technology, Steve Hunt argues, measures who is likely to get hired rather than who is likely to succeed once they do.

• Daily Pay Can Make You Less Broke. Even salaried professionals deserve earned wage access, since the company already knows exactly when you clocked in, argues Laurie Ruettimann.

• Research: AI Is Changing What Employers Want from New Hires. Jim Doucette and Vishal Gaur studied the sectors that hire the most MBAs and found employers want judgment, synthesis, and workflow redesign more than AI fluency.

• Mediabistro Weekly Drop: The Out of Office Edition. The newsletter went quiet for two weeks because its writer took a real vacation, and the takeaway is simple: whatever an AI can cover while you’re gone was already the most replaceable part of your job, per Matt Charney.

• Is HR the Most Influential Function in an Organization?. HR has the standing, Jennifer McClure argues, to influence every employee, leader, and business outcome in a company, a bigger claim than most HR departments actually back up.

• 2026 Top 100 HR Tech Influencers. HR Executive’s eighth annual list is out, with the usual familiar names and a genuinely good number of new ones this year.

• Upwork’s Future Workforce Index 2026. Skilled freelancing jumped from 28% to 38% of US knowledge workers in a year, and the freelancers making more money with AI are the ones applying judgment on top of it, not just running prompts.

• People & Progress: Changes. Kristy McCann takes on Zuck’s latest misstep and BlackRock’s big investment in the trades. Why aren’t you subscribed to this one yet?

• Welcoming Danielle Moseley to ProvenBase. After 21 years running recruiting industry events at ERE Media, my former colleague Danielle Moseley is heading to ProvenBase as VP of Revenue and Business Development. That’s a hell of a run at one company.

• Customers Don’t Separate Their Brand Experience. Kat Stupka makes the case that splitting direct mail and digital into separate campaigns and separate owners is how brands end up telling customers two different stories. I enjoyed this one, we ignore the media people can actually hold far too easily.

Have a great rest of your week!

Welcome back U.S. readers from the long holiday weekend. While you were busy celebrating the big 250 with food invented in Germany and fireworks invented in China, the federal government was gearing up to consolidate the 100+ HR systems that they are responsible for.

Hopefully it won’t take 250 years.

The Office of Personnel Management awarded a contract to Oracle to the tune of 400 or so million freedom dollars to overhaul and consolidate the HR tech mess at the federal level into one unified system. The ten-year project could eventually hit $1 billion, and I’d guess that might be very conservative.

I wrote about it all on Reworked last week.

There is, of course, drama about the award itself. Initially, Workday was picked more than a year ago via a no-bid contract. That seemed just a little suspect. Of course, Oracle is as suspect of choice with co-founder Larry Ellison on the president’s tech advisory council.

But let’s put all of that aside for a second. As hard as it may be, let’s also put aside the severe doubts I have that the OPM and Oracle can do this as well as promised. Time will tell but bringing together these systems will be painful and will have to not be torpedoed by multiple presidential administrations.

Instead, I played it straight and I offer four universal truths for HR leaders. The first one is whether a consolidation-only software thesis can truly work. Consolidation is great for system admins and government budget hawks but if the consolidation doesn’t deliver better results for agencies, it seems destined for politicized failure before it truly gets launched.

The other truths are in the post and if you’re an enterprise HR or IT leader, it’s worth a look.

And look, it is easy to clown on Oracle here for having the audacity to take on this huge task. It’s easy to clown on Oracle for being Oracle these days. They’ve let go of 20k+ people this year and blamed AI for it. They seem to not be getting hammered the same way their ERP peers are while still struggling in the same sort of ways. A project like this seems like a minefield unless you’re well connected politically on both sides of the aisle.

But I feel like that’s a cheap argument and as a U.S. taxpayer, I also feel obligated to hope that it comes in on time and under budget. Regardless of the outcome though, we’ll all get a front row seat to the show. And we can see how off the rails a consolidation of this level can really get.

What else is happening this week?

• The People Analytics Tech Market Just Hit $12.2 Billion. The market’s maturing, and Stacia Garr finds vendors unbundling consulting from the price tag as growth slows. They’re also quietly gutting the ethics guidance they used to give away free.

• Why Has Cost of Living Become The #1 Issue In The US? It’s More Complex Than Inflation.. Josh Bersin traces four decades of wage stagnation to one choice: pay productivity gains to shareholders instead of workers. Now everyone’s acting surprised nobody can afford groceries.

• Artificial Intelligence vs. Employee Wellbeing. AI wrecks employee wellbeing in seven distinct ways, Steven Hunt argues, from co-dependence to algorithmic anxiety to plain old isolation. His prescription is simple: no person should ever feel like they’re working for a machine.

• The Machine Is Only as Exponential as You. Jason Averbook digs into workslop, AI output that looks finished but isn’t, and pegs the cleanup cost at $9 million a year for a 10,000-person company. Everyone has the same AI model now, so what separates good work from garbage was never the tech.

• On Entrepreneurship, Panic Attacks, and Lightning Bolts. A fired client, a team revolt, and a panic attack in a cornfield make up Jason Seiden‘s real founder story, resolved only by a stranger’s voicemail offering to invest.

• It’s Not ‘Will AI Take Your Job.’ It’s Which Choice Your Company Already Made. Entry-level employment is down 16% in the most AI-exposed jobs, per Erik Brynjolfsson‘s research. Jess Von Bank‘s point: the question was never whether AI takes jobs, it’s which side of that choice your CFO already picked without telling anyone.

• The Hallucination of Competence. AI is a Dunning-Kruger accelerator, John Sumser argues: prose so polished we assume the thinking was ours, when a ten-word prompt did 99% of the work. His fix isn’t banning the tool, it’s reintroducing the friction that builds real expertise.

• It’s Hard To Be Principled And Employed. Laurie Ruettimann breaks down how CBS’s new owners fired a 37-year anchor for saying out loud that they were destroying 60 Minutes. Being pivotal to the most decorated show in America didn’t make him untouchable, and neither will your work.

• Introducing the Fraud Squad. “Candidate fraud is organized. Now we are too,” reads the tagline for Fraud Squad, the free community Stacy Zapar just launched for recruiters swapping notes on fake candidates.

• The Best HR & People Analytics Articles Of June 2026. The throughline of David Green‘s monthly roundup has moved from whether you’re using AI to whether anyone can prove it’s worth anything.

• The Traditional Career Ladder Doesn’t Really Exist Anymore. JD Peterson of Perceptyx breaks down the shift from change management to change capacity in this Matt Tatum interview. Really, it’s just leaders admitting employees can smell a training course from a real answer.

• I Came To This Country From Ukraine At 15. George Elfond writes about arriving in the US too scared to spend $4 on a hot dog, and the question that’s followed him since: how do you hand your kids opportunity without handing away the hunger that built it?

• Stories And People Are Everything. Brandon Giella pulled the actual data most of us are just guessing at: stories beat everything, LinkedIn beats every other channel for B2B, and reach without engagement is just noise.

• Someone Is Paying For This AI Summer. Humans training their own $2-an-hour replacements, Teradata funding its AI budget out of employee raises: Kristy McCann tallies who’s actually covering the AI bill. Her dare to CHROs: be the most human person left in the building.

• U.S. Economy Added 57,000 Jobs In June, Less Than Expected; Unemployment Rate At 4.2%. The headline unemployment rate dropped only because 507,000 people left the labor force entirely, which is a strange kind of bad good news.

• Trump Can Fire Federal Agency Heads At Will, SCOTUS Rules. The Supreme Court overturned a 90-year-old precedent 6-3, so independent agencies like the NLRB and EEOC are independent right up until whoever’s in the Oval Office decides otherwise.

Have a great rest of your week!

I am out of office this week so we have a special edition of The Turn as we look at the top work and HR tech stories of the year so far.

We’re halfway through 2026 and it’s been a helluva half year. So how do you sort through the stories?

Not every big story is actually big. Not every small story is actually small. So I attempted to measure how much a story actually moves the world of work, across six dimensions weighted by how much each one matters.

• Reach (25%) — How many workers and organizations are actually touched by this story

• Disruption (20%) — How far it breaks from the status quo

• Durability (20%) — Whether this is a structural shift or a flash in the pan

• Stakes (15%) — The dollar, market, or career magnitude involved

• Precedent (10%) — Whether it sets a legal, regulatory, or industry template

• Salience (10%) — How loud the industry conversation actually is

Each story is scored 1–10 on each dimension. The weighted scores combine into a single gravity total.

1. AI-as-restructuring layoffs go mainstream

This is the story of the year and it is not close.

Tech layoffs blew past 142,000 in 2026, led by profitable companies cutting to fund a roughly $700 billion AI infrastructure buildout. Amazon eliminated around 16,000 corporate roles. Meta cut around 8,000, about 10% of headcount. Oracle confirmed 10,000 plus. Workday trimmed 1,750 people and called it a reallocation toward AI.

AI is now cited in a majority of layoff announcements, even at companies where the actual automation is still theoretical. Oxford Economics has argued it is often cover for ordinary cost-cutting. Both things are probably true at different companies. What changed in 2026 is that "AI made us do it" became an acceptable public explanation, and companies learned fast that it was a better story than the alternatives.

On reach, this touches the entire labor market at hundreds of thousands of jobs and counting. On disruption, profitable firms cutting to fund AI breaks from the historical norm that layoffs signal distress. On durability, this looks like a structural reset of how companies think about headcount, not a one-quarter event. The stakes are enormous in human terms. And on salience, this is the dominant work-and-AI story of 2026 by a wide margin.

2. The entry-level job collapse

The other half of the AI-and-labor story, and arguably the more consequential one.

New-grad unemployment climbed to roughly 5.6-5.7% against a 4.3% national rate. Underemployment hit 42.5%, the highest since the pandemic. A Stanford Digital Economy Lab paper found a 16% relative employment decline for 22-25 year olds in the most AI-exposed occupations. Entry-level postings are down roughly 35% since early 2023.

The people this is happening to do not have LinkedIn audiences or industry contacts. They just have no way in. The career on-ramp is disappearing, and because it is slower-burning than mass layoffs, it gets less coverage. But the durability score here is high because pipeline damage compounds over years and is slow to reverse. Every year this continues is another year harder to fix.

3. Vendors pivot to a “platform of agents”

The entire HR tech category repositioned this year, from systems of record to what they are now calling systems of agents.

Josh Bersin framed it as the reinvention of Workday, and the product news backs it up. Early adopters are reporting recruiter capacity gains of 54%. Workday has promised 14 new agents by year-end and launched governance tooling including Agent Passport and Agent-Ready Tools to help organizations manage what the agents are actually doing. The Salesforce Agentforce partnership adds another layer to the platform play.

The reach here is significant: eventually this touches every organization running a major HCM suite, and through them most knowledge workers. The disruption score is high because repositioning HR tech from record-keeping to autonomous work execution is a fundamental change to what the category does. And the precedent score reflects that governance tooling like Agent Passport represents an early template for how organizations deploy and oversee AI agents at work. Whether the market is right that digital workers become the next platform layer is still the question. That every major vendor is making some version of this bet is not.

4. Mobley v. Workday sets the AI-hiring precedent

The first real legal template for algorithmic hiring liability.

In a landmark March 6 ruling, a federal court rejected Workday's argument that the ADEA does not cover applicants screened by a third-party system, letting the AI age-discrimination collective action proceed. The opt-in window closed March 7.

What this case is building is the first real legal framework for who is liable when an algorithm filters out a candidate. That answer matters for every vendor selling AI screening tools and every employer running them. The precedent score is the highest of any story on this list for that reason. Case law has long downstream effects, and this case defines the terrain for algorithmic hiring liability in a way nothing before it has. The reach is broad but indirect: any organization using AI in hiring has a stake in the outcome, even if they are not named in the suit.

5. Return-to-office hardens to five days

The hybrid detente is over at the top of the market.

55% of Fortune 100 companies now require five-day attendance, up from 5% in 2021. The federal government mandated full return. Roughly 47% of strict-RTO firms say they will discipline non-compliance. Actual measured attendance has barely moved.

The reach here is enormous: this affects huge swaths of office workers and the entire federal workforce. But disruption and durability scores are more moderate, because this is an escalation of an existing fight rather than a new rupture. The hybrid conversation has been going on for four years. What changed in 2026 is that major employers stopped leaving room for negotiation. Whether that holds as the labor market shifts is a different question. For now the mandate went one direction and the behavior went another, and that gap is the whole story.

6. The EU AI Act’s employment rules land in limbo

Regulatory whiplash for anyone building or buying hiring AI in Europe.

High-risk employment AI obligations were set to apply August 2, 2026. Then a May 7 provisional Digital Omnibus deal proposed deferring them to December 2027, with the original timeline still in force if the deferral is not formally adopted in time. Fines reach up to 15 million euros or 3% of global turnover.

The uncertainty itself is the story. Compliance teams are now planning against two possible timelines simultaneously. The deferral was supposed to reduce pressure. The result is more complexity, not less. The precedent score here is high because the EU AI Act is the global reference point for regulating employment AI, and what happens with the enforcement timeline will shape how other jurisdictions approach the same question. The framework will govern compliance for years regardless of when the clock starts.

7. The Deel-Rippling espionage war escalates

The juiciest saga in HR tech kept escalating all year.

Federal rulings in February and March let Rippling's RICO and trade-secret claims advance while letting some Deel executives off as individual defendants. Deel filed counterclaims. New banking records surfaced that Rippling says prove Deel paid the person who stole from them. And through all of it, Deel raised $300 million at a $17.3 billion valuation, higher than the year before.

The direct reach is mostly limited to the two companies and their customers. But the stakes are significant: two companies valued around $17 billion each, RICO and trade-secret exposure, and a fundraise that proves the market is not particularly bothered by any of it. The durability score reflects that litigation and reputational effects will run for years. The salience score is the highest of any story in the bottom half of the list. This one has been impossible to ignore.

8. Dayforce taken private

The year’s biggest HCM deal.

Thoma Bravo agreed to acquire Dayforce for roughly $12.3 billion including debt, with a minority investment from the Abu Dhabi Investment Authority. The deal is explicitly framed as a bet on expanding Dayforce's AI capabilities in an environment where public market scrutiny makes long-term investment harder to execute.

It ranks eighth. A $12.3 billion acquisition ranks below a lawsuit with no settlement, a mandate nobody is following, a regulatory deadline in limbo, and a corporate espionage saga. That is what happens when you weight worker impact over dollar amount. The stakes score is high, but reach and disruption are moderate: this affects Dayforce customers and the vendor landscape, not workers broadly. PE consolidating HCM is not a new dynamic even at this size. The immediate experience for the average Dayforce user does not change.

9. The FTC noncompete rule is officially dead

The sweeping federal ban is over.

The FTC formally removed the Non-Compete Rule from federal regulations on February 12 and shifted to case-by-case Section 5 enforcement. A subsequent action against Rollins covered around 18,000 workers, which gives a sense of the scale case-by-case enforcement operates at.

What replaced a national ban is a growing state patchwork that is in some ways messier to navigate than either clean outcome would have been. The workers noncompetes affect most are still subject to them in most states. The reach score reflects that noncompetes touch a large share of the workforce. But disruption scores lower because this restores the prior status quo rather than breaking new ground. The headline moved considerably further than the actual situation did.

10. Mercor and the “train your replacement” model

A genuinely evolving trend for recruiting and knowledge work.

Mercor, now valued around $10 billion, pays professionals hourly to teach their workflows to AI agents and runs an AI-interview-first hiring process for new roles. The company generated significant attention after Bloomberg coverage in the spring.

The reach score is low today: one company and a niche labor pool. But the disruption and precedent scores reflect what this model represents. Paying people to make themselves replaceable as a product built on top of a hiring platform is a structure that has not existed before. The durability score is moderate because it is still unproven at scale. This earns its spot as a signal of where AI-native work models may be heading, not as a story with current mass impact.

What are your top stories for the year so far? What do you think will happen in the rest of the year?

When I heard that Uber had blown through their whole yearly AI budget by April because they promoted tokenmaxxing (like many of their peers), I thought what I usually think when tech companies do something predictably dumb: Why is anybody surprised?

About a month later, the rideshare company let go of nearly a quarter of their HR team.

Okay, that was a little surprising.

Officially, Uber has denied any connection between these two events. But I have my doubts so I wrote about that in my Reworked column this week.

According to Uber leadership, HR was, “too complex and fragmented, with overlapping responsibilities, unclear ownership and teams operating too far from the businesses and partners they support.”

And according to reporting from HR Executive, of those remaining in the HR department, anyone still on a remote work arrangement is having it rescinded.

Here’s a guess that HR won’t be done with losing people.

Like so many tech companies, Uber is not struggling financially. Its first quarter gross bookings were up 25% and its YoY revenue is up nearly 15%. Of course, the stock market continues to not care about that — its share price is down more than 13% for the year as of this writing.

So they did what a lot of companies do: a very loud layoff of corporate’s favorite scape goat with the hope to appease the shareholder class.

Oh, we’re back to not being surprised again.

I am not a huge talker but I have a whole rant about the types of companies that throw HR under the bus like this. Like any HR department ever gets the benefit of the doubt when it comes to increased staffing in their department. Do I believe Uber was overstaffed by 340 HR people? No, I do not.

Do you know how many HR leaders I have talked to who have had to write multi-page business cases for the most ridiculous things? A six-month contract recruiter to cover for maternity leave? A single HR analytics employee when the business has asked for dozens of detailed reports from ancient enterprise systems?

Maybe HR did need a re-org. But the timing, the messaging, and the impact don’t add up for me. HR departments getting cut this significantly are for flailing businesses or mergers, not for a some complexities and overlap in a department.

In my column, my hope is that HR increases their presence in discussions about AI usage across the organization (not just for HR). I keyed in on three discussions HR deserves to be a part of. Take a read if you’re interested.

What else is happening this week?

• What Olivia Rodrigo and Robert Smith Can Teach Us About Enterprise Software Today. A Thomas Otter post with a music observation built in is two of my favorite things at once. He uses the Rodrigo-Smith Glastonbury collab to argue that enterprise incumbents can either coast on back-catalog nostalgia or take real risks and earn new audiences.

• Qualified, Interested, and Affordable Is No Longer a Valid Method for Employers to Measure Applicants. Time to QUAIL. Andrew Gadomski adds two letters to the old QIA framework, Understandable and Legitimate, to account for a world where fake candidates, fake recruiters, and fake job postings mean neither side of hiring trusts the other anymore.

• Crash Out With Me: A Letter to Corporate. Madison Butler published this anonymously because the pattern matters more than the details, and any Black woman who has tried to earn safety inside systems not designed to provide it will recognize it immediately.

• You Don’t Have to Say Yes to Money. Laurie Ruettimann on the Peter Thiel Dialog list and the prominent leadership authors on it: people who make a career preaching integrity should know who is signing the check.

• Pro-Worker AI, Explained. MIT economists cover the difference between AI that replaces labor and AI that expands what workers can do, and why most companies are defaulting to the easier path.

  The Work AI Can’t Do. Moe Carrick on connective labor: the unscheduled conversation, the noticed silence, and the manager who actually knows her people.

• What Jalen Brunson Can Teach Us About Team Building. David Manaster at ERE is with me on this: Brunson left money on the table so the Knicks could build a real roster around him, and they won the chip. You have to do a lot of things right, and it all starts with people.

• Is DEI Dead? Not According to New Catalyst Data on Workplace Inclusion. New data shows 80% of organizations remain committed to DEI, and that 55% have publicly signaled a retreat while only 34% have actually reduced their inclusion efforts. Sorry not sorry: still kicking.

• Toxic Workplaces Are Winning—For Now. New Businessolver data shows that the worst corporate cultures are 2x more likely to report significant financial growth. The pendulum will swing the other way… eventually.

• Ragan’s 2026 HR Tech Hot List Honorees Announced. Twenty companies on the list this year, some new names alongside the familiar ones like UKG, Rippling, Lattice, and Eightfold.

• How Remote Work Has Helped a Generation of Working Parents. Mothers of children under 5 have been in the labor force at higher rates since 2023 than before the pandemic, and some say they wouldn’t have had kids at all without remote flexibility.

• Growth of Data Centers and AI Expose Tensions in Organized Labor. Building trades want the construction jobs data centers bring; other unions are worried about AI taking theirs. The reality of diverging economic incentives can be brutal for working folks.

• Newsom Promised to Help Californians Build New Careers. Now, the Money Is Running Out. California’s high road training partnerships are getting little or no new funding in the 2026-27 budget. Promising a workforce agenda and not funding it is just a press release.

• Fake Candidate? Recruiting Impostor? Check Here. Stacy Zapar built a free, searchable database of suspicious emails, profiles, and domains used by scammers: 1,200+ flagged addresses and 170+ recruiter impersonators. Bookmark it.

• Now of Work Weekly Digital Meetup with Jason Lauritsen. If you haven’t tuned into Jason Averbook and Jess Von Bank‘s weekly show, catch up on this one with Jason Lauritsen and sign up to join on Fridays.

• When the Prediction Becomes the Verdict. Part two of Jess Von Bank‘s seven-part series: the Kentucky pretrial risk assessment tool deployed to reduce racial bias ended up giving racism a more defensible interface. Your AI governance team should read this one.

• Another Word for Technical Debt Is Maintenance. John Sumser on agents: the harness is 90% of the model, ages on a different clock than the LLM, and when the model improves, the guardrails built around its old blind spots quietly stop working. “Agent reliability engineer” might be the next job title nobody saw coming.

Make your writing more interesting

Last week, I shared a short guide about how to fix your brand writing. Of course, I had to promote myself at the end, but even if you don’t want to work with me, there are a few takeaways to make your writing stand out.

I actually don’t blame AI for all of the terrible writing you see from B2B brands. This has been happening for years before AI could write an em dash. Whether you use AI or not, there are some pretty simple fixes, though.

Swipe through the guide here, typo and all because I decided to push publish instead of sitting on it for days.

Have a great rest of your week!

Every time I think about Orlando, I think about the band O-Town and their 2001 hit All or Nothing. They were formed on the TV show Making the Band, created by the same guy who formed the Backstreet Boys and NSYNC.

(This was the pre-Diddy version of Making the Band, btw. I’m not looking to get cancelled.)

It’s not good music but when you are 19 and all the girls you liked watched it and sang along to it, well, now I know it for probably the rest of my life.

The earworm of that overproduced single came back to me as I saw folks online descending into Orlando for SHRM’s annual conference.

The first annual conference I attended was in New Orleans in 2009. It was tiny by SHRM standards but still pretty damn big. Something along the lines of 7,000 or so attendees. I met so many people. And as a credit to them, I talked on a panel on about using social media as an HR professional. They even paid for travel. Shortly thereafter, I lost my HR job for good and made the pivot into whatever this is.

I also have so many good stories about that trip that I won’t publish because walking to my hotel barefoot came after some memories that are now seared into my eyeballs.

I’ve been writing about SHRM for two decades now, though. I started my blog, Your HR Guy, in May 2006 and wrote about missing the annual conference then. My last post about SHRM was three days before Christmas. In between has been a rollercoaster of (a few) ups and (more) downs.

I get annoyed at the world’s largest HR association for a lot of (I think) valid reasons. They have lobbied on the profession’s behalf to make the laws that protect people in the workplace less effective. They sometimes act as if Toby’s flaw in the sitcom The Office was that he didn’t side with management more.

It also seems like not a great place to work. In addition to the two lawsuits I wrote about late last year, the whole “come to the office or we’ll send your job overseas” thing wasn’t great. As “The Voice of All Things Work,” they should at least be specific about what kind of workplace they want to represent.

As I’ve spent the last two decades picking at SHRM, they have only become bigger and more influential. There will likely be something on the order of 25,000 people at the show this week. Their membership has grown. The CEO has a bigger microphone than ever.

And personally, friends and colleagues are more intertwined with the organization too. They make hay at the conference. They publish books with them. They speak there. They write for them. They advise for them.

I get it. Some of them are fans and others are making the tradeoffs you have to make to pay the mortgage. Yet, I still can’t get behind SHRM. All or nothing, I guess. And maybe I’m worse off for it.

The great thing is that there are so many more communities for folks who want an alternative. And honestly, I’m a little jealous because my former HR career had very few other options. But I might not have started writing which means I wouldn’t be here writing to you.

So it worked out, and I am mostly at peace with the all or nothing part. I can only wish the SHRM attendees in my life the best time as you can have in Orlando. Good luck out there, learn something, make some connections or money, and come back refreshed.

And a quick Orlando airport tip: If you can’t get access to a lounge for some peace from all of the loud kids, the smoking lounge is quiet and probably only knocks a few days off your life. They would’ve sucked anyway.

What else is going on this week

• 80% Cut Jobs for AI but Got No ROI: Gartner Study. Gartner surveyed 350 executives at billion-dollar companies and found that cutting headcount to fund AI didn’t move the returns needle. The companies that saw gains used AI to amplify people, not replace them.

• The AI Outcomes Gap Is A Waiting Problem. Business leaders have decided the technology will get better and solve the outcomes problem on its own, so they’re waiting instead of changing culture, upskilling people, or measuring anything, argues Gabby Burlacu. That bet is not going well.

• The AI Layoff Wave Is Becoming a Powder Keg. AI insiders are minting generational wealth in the same quarter tens of thousands of workers are hitting the door, and Connie Loizos notes the historical precedent for what happens when that gap gets wide enough.

• What It Felt Like to Be an Applicant. Nearly two decades helping build the HR tech industry, then Shannon Pritchett entered the job market and found the whole thing remarkably broken even with her level of access and relationships.

• More Than Half of Young US Hospitality Workers Would Give Up 5% Pay Raise to Feel More Confident. New research from Attensi finds younger hospitality workers would trade a raise for training that actually builds confidence. Two-thirds said current training delivers information instead of practice, which is a polite way of saying it doesn’t work.

• Tech Layoffs 2026 Hit 149,935: Uber Cuts HR Days After AI Drained Its Coding Budget. Uber burned through its entire 2026 AI coding budget in four months, with nearly 70% of committed code now AI-generated, then cut 23% of its HR division. The company insists the two facts are unrelated.

• Amazon Warehouse Workers Allegedly Asked for a Raise and Were Given a Bounce House Instead. Amazon upgraded the pizza party to a bounce house when workers asked for higher wages. Thanks?

• Tollbooth. Indeed made sponsored listings the default and buried free postings. Brian Fink has the apt metaphor: a road everyone assumed was public just got fenced.

• Very Dry but Necessary. AI governance is prospective where financial governance is retrospective, and the people good at defining constraints are almost never the people doing the discovering, writes John Sumser.

• Ask the Analyst: What Are Best-Performing Organizations Doing to Move Beyond Traditional Best Practices? The best organizations ask what problem they’re solving before they ask what everyone else is doing, according to Nicole Roberts, Nehal Nangia, and David Perring. Turns out copying what worked somewhere else at a different growth stage is mostly how you get trapped.

• Sorry, Your Job Is Awesome! Laurie Ruettimann is launching a limited-run podcast for people who actually like their jobs and want to get better at them, and this is a great idea. There’s no shortage of content about broken workplaces.

• Your AI Content Problem Is Actually a Data Problem. The blandification problem isn’t a writing problem, it’s a missing inputs problem, argues Steve Smith. The shortest stave in the barrel is almost always real customer voice, and no amount of prompting fixes that.

• It’s Not an AI Problem. It’s a Bad Idea Problem. My post isn’t so sure this is an AI problem or a bad idea problem. A lot of brand writing was already bad before the models showed up.

• The House Always Trades Up. McDonald’s premiumization, Vegas losing its middle-class customer, employers hollowing out the dependable middle of the workforce: Mike Wood connects all three. The person being priced out of the drive-thru and the person being automated out of the job are the same person.

• 25 Days Off Wasn’t Enough to Recover From Burnout. Surgery, burnout, a lifetime of working since 16, and the finding that time off alone doesn’t fix it: Kat Kibben writes about all of it honestly.

• The Titanic in Plain Sight. AI deployment decisions and their workforce consequences are being governed by different people on different timelines with no one accountable for the gap, opens Jess Von Bank in a seven-part series. Subscribe if you aren’t already so you can read the whole thing.

• Making Sense of the Labor Numbers. The labor headlines and what’s actually behind them, from Mel Plett and Francesca Ranieri in a new Your Work Friends episode.

• Introducing The Humanity in AI Awards. Sounds like an oxymoron, but Beth White is working to make sure the people on the other side of tokenmaxxing are recognized too.

• Human Capital Insider: June 2026. Kyle Forrest gives his monthly rundown of Gartner research and what’s moving in human capital strategy.

• Advocates Release New Report Urging Congress to Restore Federal Contractor Employees’ Rights After Trump’s Rescission of E.O. 11246. The right message aimed at the wrong audience, given the current composition of Congress.

• Oracle Wins $396M Federal HR Systems Overhaul Contract. OPM picked Oracle to consolidate more than 100 HR systems covering two million federal employees on a 10-year contract. Set your alarms for a decade from now.

• A Farewell to the Pay Transparency Newsletter. After three years, Anita Lettink is wrapping up her Equal Pay newsletter and moving to her Pay Time! newsletter. Subscribe there and follow her work.

• John Wilson Steps Down After 20 Years Leading Wilson HCM. Happy trails to John Wilson, one of the good ones. The kind of exit that generates comments like this doesn’t happen by accident.

Have a great rest of your week!