Artificially Intelligent, Actually Useful.

A Weekly AI Update So You Don’t Have To Wing It

View another: Elsewhere

Jul 4, 2026 - 06:00 UTC

February 25, 2026

Artificially Intelligent, Actually Useful.

A Weekly AI Update So You Don't Have To Wing It

The Tale of AI and the White-Collar Job

This week's newsletter is about two seemingly contradictory stats. Read the news and find that AI is gutting white-collar work. Read the monthly jobs report from the Bureau of Labor Statistics and find that the unemployment rate has barely moved. Both come from real data, and both are right.

The contradiction resolves once you stop looking at the top-line number. The unemployment rate holds because losses and gains are canceling out. Routine, entry-level roles have been thinning out. But the roles that reward discernment, judgment, and expertise are growing fast enough to absorb the difference.

This issue follows the story in three ways: what the big layoff numbers actually signal, why some of those same companies are already rehiring, and how entry-level roles are being rewritten.

Layoffs as Investor Theater

Cloudflare cut 1,100 jobs, calling it AI strategy, while reporting record quarterly revenue of $639.8 million. Meta eliminated 8,000 roles while posting $26.8 billion in net income. Block slashed 40% of its workforce while projecting $12 billion in gross profit. These are not struggling companies.

Gartner studied companies deploying AI at scale and found the ones cutting staff were not seeing better returns. Trimming payroll frees up budget. It does not, on its own, create returns. Deploying enterprise-level AI does not replace judgment-heavy work.

Key Insight: These massive, newsworthy layoffs are not proof that AI will replace jobs. They are proof that AI at scale is expensive and investors need to see cuts somewhere. The missing returns and need for critical thinking are the counterpoint.

Read the Forbes Analysis →

The Changes in Entry-Level Work

PwC's 2026 Global AI Jobs Barometer finds that entry-level work is not emptying out; it is being rewritten. In the most AI-exposed occupations, junior roles are now seven times more likely to demand senior skills like leadership, judgment, and creativity. Openings for these 'seniorized' roles grew 35% since 2019, while other entry-level roles shrank 10%. The same divide runs through the wider market: roles where AI amplifies human expertise show twice the job growth and 42% faster salary growth than roles where AI makes the work easy for non-experts.

The Bottom Line: The routine work that used to be the on-ramp for entry-level workers is the first to be automated. The jobs are still there; they're just asking newcomers to arrive with experience.

See the PwC Barometer →

The Impact Nobody Saw Coming

Benedict Evans makes a blunt case: scoring which jobs AI will hit is predicting what cannot be predicted, and the back-test proves it. Accounting should have been the first job computers wiped out. We built machines to create spreadsheets, reports, and even ERPs, yet the number of accountants kept climbing. Taxi driving seemed untouchable by 'internet exposure,' yet cellphones and location data produced Uber and turned million-dollar medallions worthless.

Key Takeaway: Treat every AI-exposure chart as a guess, not a plan. Bet on people who adapt and on workflows you can rewire fast, and ask what does AI make less expensive two steps upstream from you.

Read his Breakdown →

Quick Hits

.Foundations

AI State of Affairs

A senior editor at MIT Technology Review discusses AI in 2026. On jobs, there is still almost no hard data on what will happen to employment. The dystopian future isn't here, but near-term risks people warned about have started to arrive. The backlash is organizing into real protests. His grounding line: this is a marathon, not a sprint.

.Video

What Jobs Data Says

The CEO of Europe's largest freelance marketplace says that demand for agentic AI skills grew 60x this past year. Hiring graduates is also picking back up after an early freeze (a sign that the fear of mass displacement got ahead of reality). The shift is in the work itself, not in whether the work exists.

.Deep Dive

A Tiered Playbook for AI's Potential Labor Shock

This detailed economic policy framework from Anthropic ties government response to the unemployment rate itself: staying idle at low disruption and expanding support as the numbers worsen. The most useful line for any employer: decide in advance how you will use freed-up capacity instead of defaulting to headcount cuts.

Industry Developments

Adobe and LinkedIn Bet on Marketer Reskilling

A free training for marketers was launched by Adobe and LinkedIn, citing a 113% year-over-year jump in job postings requiring AI literacy. The courses are role-based and available in 47 languages. Free, credentialed AI training from a platform this size raises the floor for what counts as "AI literate."

Midjourney Builds a Body Scanner

Midjourney, the AI image-generation company, announced a full-body ultrasound scanner it claims rivals an MRI with no radiation and a 60-second scan, built on Butterfly Network's chips under a co-development deal worth up to $74 million.

This is the other side of the jobs story. A product category that didn't exist last year implies new jobs ranging from on-device imaging engineers and scan operators to the developers building diagnostic apps on top of the platform. New tools spawn new roles, and those rarely show up in any exposure forecast.

Alright, that's a wrap for the week!

So, where is AI going, and which jobs are safe? I wrote this issue because that's a question I get on a regular basis. The honest answer is that nobody knows. Some work is vanishing while other work is becoming more valuable. And although the specifics are impossible to predict, the pattern of displacement and creation is historically well documented.

One part of this story that had me concerned was the promotion track. If junior roles are being automated away, how will companies staff the senior teams needed in five to ten years? The encouraging news is that some companies are starting to rebuild that rung rather than remove it.

If you're staffing for the next two years, I'd love to know how you're thinking about entry-level roles. Are you redesigning them, freezing them, or hoping the question resolves itself? And the same question goes for those of you at staffing companies: what are you seeing from your clients?

Thoughts?

Artificially Intelligent, Actually Useful. is brought to you by iS2 Digital
Your Software at the Speed of Change

Click here to subscribe, if you want to receive our newsletter directly.

Artificially Intelligent, Actually Useful.

Narrowly Brilliant

LLMs Are Growing Up

AI Isn't What You Think It is

Context is King

Artificially Intelligent, Actually Useful.

Layoffs as Investor Theater

The Changes in Entry-Level Work

The Impact Nobody Saw Coming

Quick Hits

Industry Developments

Got a project in mind?
Tell us about it.

Artificially Intelligent, Actually Useful.

Narrowly Brilliant

LLMs Are Growing Up

AI Isn't What You Think It is

Context is King

Artificially Intelligent, Actually Useful.

Layoffs as Investor Theater

The Changes in Entry-Level Work

The Impact Nobody Saw Coming

Quick Hits

Industry Developments

Better Business, Not Better Buzzwords

Got a project in mind?Tell us about it.

Got a project in mind?
Tell us about it.