Three things lined up this week. The S&P 500 closed at a new record. Healthcare added 82,000 jobs, nearly half of all hiring. A peer-reviewed study tracked 160 companies that tried the 4-day work week, and 90% of them kept it.

Also: Snap's CEO published the first real number on how much of their code is now AI-generated (65%). That number gives you something you haven't had until now, a benchmark.

Let's get into it.

Anthropic and OpenAI together just crossed $44B in annualized revenue. Anthropic hit $19B ARR this week, up from $9B in January. OpenAI is at $25B. Between them, they now have over 7,000 open roles posted this quarter, from ML engineers to customer success to policy. The AI boom isn't only productivity gains at other companies. It's also the single biggest hiring story of 2026.

Healthcare added 82,000 jobs in March, nearly half of all net hiring. BLS's March breakout confirms healthcare is the part of the labor market still running hot. Nurses, NPs, health services managers. If you've been wondering where demand isn't softening, here it is, with five-plus years of demographic tailwind behind it.

A Nature study tracked 160 companies that tried the 4-day work week. 90% kept it. Published this week. Productivity held or improved at 86% of participating companies, employee burnout dropped 41%, revenue held. Ten months later, only 10% reverted. The data is getting hard to argue with, and the list of companies piloting it keeps growing.

The S&P 500 closed at a record 7,022.95 yesterday. The 14th record close of 2026, driven by strong Q1 bank earnings (JPMorgan beat on EPS, fixed income trading up 21%) plus the AI infrastructure wave. Records matter for more than sentiment: household net worth just hit an all-time high, which typically supports hiring two to three quarters out.

Disney cut 1,000 corporate jobs this week. Announced April 14, first under the new CEO. Internal memos describe it as an AI-driven restructuring of creative operations. Disney joins Amazon, Oracle, and Snap in the Q2 layoff wave. If you work in a role where AI tools are now standard issue, the pattern is getting explicit enough that you can position around it. The Big Story has the move.

Another headline. Another client pays late. The next 10 days shift. You open your bank app before walking into the office.

The hits just keep coming right now.

And as the leader, you’re the one absorbing all of them.

But survival doesn’t come from holding tighter alone.

The Small Business Survivor Guide gives you 83 practical ways to cut costs, stabilize cash flow, and navigate economic pressure with confidence.

Because in times like these, stability isn’t luck. It’s strategy.

And the leaders who stay standing are the ones who prepare for what’s next.

Download The Small Business Survival Guide

On Tuesday, Snap's CEO told staff that close to two-thirds of new code at the company now ships through AI tooling. Same day, the company cut 500 engineers. The two announcements came from the same office on the same afternoon. That wasn't an accident, and it's also not actually bad news if you know what to do with it.

Here's why the number matters. For the last two years, "AI productivity" has been a soft story. Companies gestured at it on earnings calls. Analysts estimated it. Nobody had actually published a ratio. Now someone has. And a specific number is something you can benchmark against.

If you're currently in a role where AI tools are available, the question is no longer "will AI change my work." It has, and companies are being explicit about it. The question is "am I on the right side of the 65%?" Meaning: are you the person making an AI tool 10x more useful than it is on its own? Or is your output roughly at or below what a well-prompted model produces without supervision?

The people who come out of this cycle ahead are the ones who can show two things. First, that their AI-assisted output is measurably better than what AI produces unaided. Second, that they can document the delta. Save your best prompts. Keep before-and-after examples. Note where your judgment moved a piece of work from a B to an A. This is the next version of "portfolio" or "deal sheet," and almost no one has built one yet.

Worth noting: Snap's engineering team isn't being wiped out. 84% of engineers still have their jobs, and they were given the same AI tools that produced the 65% number. Many will be getting raises, because the ones who win are the ones who learn to use the tools to run circles around anyone who hasn't.

Why it matters: A specific number means you can set a target. Pick one part of your work where AI could plausibly cover 65% of the task. Then make your human contribution on the other 35% so obviously better that the ratio becomes the case for paying you more, not less. That's a raise conversation, a resume line, and a case study all at once.

LinkedIn dropped its 2026 Jobs on the Rise report a few weeks ago, and the top of the list tells you where the market is actually spending its hiring budget right now.

Three of the top five fastest-growing roles in the US are AI-focused: Artificial Intelligence Engineers, AI Consultants and Strategists, and Data Annotators. That's a big share of the top tier going to roles that barely existed four years ago. Healthcare and skilled trades fill out the rest.

The interesting part is what's sitting at #7 and #9. Strategic Advisors and Independent Consultants landed at #7. Founders landed at #9. These are self-employed categories. LinkedIn's own data is flagging a shift: the fastest-growing "jobs" include a bunch of roles where you work for yourself.

Put those two signals together. The fastest-growing skill on the platform is AI, and the fastest-growing employment category is self-employed. That's the real story. Enterprises are consolidating headcount while buying AI expertise, and a meaningful chunk of that expertise is being sold back to them by independents.

Healthcare still matters too. Nurse Practitioners are projected to grow about 40% between 2024 and 2034, driven by aging populations and team-based care models. Not every career move is about code.

Why it matters: If you're reading job market signals and only paying attention to layoff headlines, you're missing half the picture. Two kinds of careers are growing: roles deeply involved in AI deployment, and roles where you package expertise and sell it independently. Everything else is getting compressed.

The playbook most people are quietly running is to acquire AI-adjacent skills inside a job, then take those skills out as a consultant or advisor. That's not a bet on the future. It's what the top 10 LinkedIn categories say is already happening.

With pay transparency laws now covering 46% of the US workforce, you don't have to guess anymore.

If you want to get fancy: Salary.com's Personal Salary Report gives you a free comparable based on your location and role in about 5 minutes, and cites the data it uses.

Snap just became the first major company to put a hard number (65%) on how much of their code is AI-generated. Now that one CEO has done it, others will feel permission to follow. By June 30:

Last issue's PPI prediction: March PPI came in hot at 0.8% MoM, which was the "higher than February" call. Markets shrugged it off and made a new all-time high the next day, which tells you something about the current regime.

NASA cleared Artemis II for an October launch after the final integrated systems review, putting the first humans back around the Moon since 1972. (NASA)

A genetic variant that protects against preeclampsia was just identified, opening the door to a screening test within 3-5 years. Preeclampsia kills roughly 76,000 women globally each year. (Reuters)

Beaver-built wetlands across the UK are absorbing three times more carbon than models predicted, making the case for rewilding meaningfully stronger. (New Scientist)