• Everyday AI
  • Posts
  • Ep 730: Is AI creating a great recession for white collar workers? Inside Anthropic’s labor report

Ep 730: Is AI creating a great recession for white collar workers? Inside Anthropic’s labor report

Meta Picks Up Moltbook, Yann LeCun launches AMI, Google Adds Gemini AI to Workspace Tools, and more

Author

Everyday AI
March 10, 2026

 

Outsmart The Future

Today in Everyday AI
8 minute read

🎙 Daily Podcast Episode: Is AI creating a great recession for white-collar workers? A new Anthropic report suggests something more complicated is happening. Give today’s show a watch/read/listen to find out.

🕵️‍♂️ Fresh Finds: NVIDIA Launches the NemoClaw AI Agent Platform, Anthropic’s pricey code reviews, OpenAI Delays ChatGPT’s Adult Mode, and more Read on for Fresh Finds.

🗞 Byte Sized Daily AI News: Meta Picks Up Moltbook, Yann LeCun launches AMI, Google Adds Gemini AI to Workspace Tools, and more. Read on for Byte Sized News.

💪 Leverage AI: Anthropic analyzed millions of Claude interactions and found most companies are using only a fraction of what AI can already automate. Keep reading for that!

↩️ Don’t miss out: Miss our last newsletter? We covered: Microsoft shocks with Anthropic-backed Cowork release, Anthropic officially sues government, Stargate expansion hits roadblocks and more. Check it here!

Ep 730: Is AI creating a great recession for white collar workers? Inside Anthropic’s labor report


AI hasn't caused mass unemployment. Yet. 😳

Although Anthropic's new study on the labor market impacts of AI showed no real signs of massive job disruption in jobs exposed to AI, it did outline which sectors and job types might be first on the AI automation chopping block.

On today's show, we'll not only break that down, but also dive deeper into a more immediately impactful job trend that AI has caused. And your company definitely can't afford to ignore it.

Also on the pod today:

• AI's secret job apocalypse? 🤖 
• 14% drop: young job hires 📉 
• Quiet hiring vs. quiet quitting 🕵️

 

It’ll be worth your 28 minutes:

Listen on our site:

Click to listen

Subscribe and listen on your favorite podcast platform

Listen on:

Here’s our favorite AI finds from across the web:

New AI Tool Spotlight – Visual Translate automatically detects, erases, and translates on-screen text in videos, Chronicle is designed for teams making business criticalpresentations, giving you AI superpowers without sacrificingflexibility and taste, Spine Swarm is the world's first truly agentic platform designed to autonomously manage and orchestrate AI agents.

Claude Code Bugs — Anthropic’s AI now spots more code bugs than humans, with automated reviews costing up to $25 per pull request. Find out if it’s worth it.

AI or ICE — Nearly half of Americans view AI more negatively than Trump or ICE, according to a new NBC poll. Learn More

Nvidia Plans NemoClaw — Nvidia is rolling out Nemoclaw, an open-source AI agent platform designed for businesses. Find out how this could change enterprise AI development.

Top 100 AI Apps — A16z just dropped their list of the top 100 gen AI consumer apps. Check out which tools are actually taking off with real users.

Helix 02 Demonstration — Figure’s Helix 02 robot just showed off a full living room cleanup, all on its own.

Luma AI Tops Charts — Luma’s Uni-1 just beat Google and OpenAI at visual logic. Curious how it pulls off wild image combos?

Copliot Health — People are using Copilot for personal health advice more than ever, especially when it matters most. See what’s driving this late-night health trend.

ChatGPT Adult Mode Delayed — OpenAI is hitting pause on ChatGPT’s “adult mode” to boost performance and safety first. Want to know what’s next?

1. Meta Makes a Move: Acquires Moltbook AI Network 🤯

Meta has just announced its acquisition of Moltbook, an AI-powered social network, marking a significant shake-up in the tech landscape.

The deal comes at a pivotal moment as major players race to integrate next-generation AI agents into social platforms. Moltbook’s technology promises to enhance user interactions and automate community management, positioning Meta to lead the charge in smarter, more personalized digital experiences.

2. OpenAI Snaps Up Promptfoo to Boost AI Security 🛡️

OpenAI just announced it’s acquiring Promptfoo, a fast-growing AI security platform trusted by Fortune 500 giants, to supercharge its Frontier platform for enterprise AI.

The move puts OpenAI at the forefront of agentic security testing, baking automated risk detection and red-teaming directly into development workflows. With Promptfoo’s tech and team joining forces, OpenAI aims to streamline oversight, compliance, and reliability as businesses roll out AI coworkers.

3. Anthropic Launches Deep AI Code Review for PRs 💻

Anthropic has rolled out a new AI-powered Code Review tool, now in beta for Team and Enterprise users, aiming to tackle the bottleneck of rushed software reviews.

The system deploys multiple agents on each pull request, promising thorough bug detection that outpaces traditional skim reads. With average reviews taking around 20 minutes and catching issues even human reviewers miss, Anthropic says the tool is already boosting substantive feedback on PRs.

4. Google Adds Gemini AI Upgrades to Workspace Apps 📈

Today, Google launched new Gemini AI features for Docs, Sheets, Slides, and Drive, available first to Ultra and Pro subscribers. Users can now draft documents, build spreadsheets, and create presentations with AI that pulls info from their files, emails, and the web.

The update also brings smarter search and instant summaries in Drive, aiming to save time and streamline workflow. These upgrades are rolling out in beta now, with more languages coming soon.

5. Pentagon Taps Google AI to Streamline Daily Tasks ⚒️

In a fresh move to modernize its workflow, the Pentagon is rolling out Google’s Gemini AI agents for roughly 3 million employees, aiming to automate routine, non-classified tasks.

According to PANews, the initiative is currently limited to unclassified networks but discussions are underway to expand these smart agents into more secure, classified systems. This shift comes as Google doubles down on making AI tools accessible for both military and civilian staff, despite its rocky history with previous defense collaborations.

6. Yann LeCun’s AMI Raises $1B for Next-Gen AI Models 🌍

Paris-based startup Advanced Machine Intelligence (AMI), cofounded by Meta’s former chief AI scientist Yann LeCun, just scored over $1 billion in funding to push AI beyond language models and into world modeling. The company, now valued at $3.5 billion, aims to build AI systems that understand the physical world and reason like humans, challenging the industry’s current obsession with large language models.

LeCun, a major voice in AI, argues that true intelligence needs more than just words and that AMI’s approach could reshape how AI is used across manufacturing, robotics, and biomedical fields. AMI’s launch marks LeCun’s first commercial venture since leaving Meta, and he’s betting big that world models are the future of artificial intelligence.

Anthropic just matched millions of Claude conversations with U.S. job market data and found something most leaders are reading completely wrong.

Right now, data does not support AI causing mass unemployment.
So does that mean AI is not really coming for jobs? Nah. It means companies still do not understand what AI can already automate.


And sorry... if your team saw “no mass unemployment” and relaxed, you are prolly celebrating the wrong headline.

That is the real punchline from the new Anthropic labor report. Researchers mapped millions of anonymized Claude chats against 20,000-plus work tasks across roughly 800 occupations using federal job data, and the loudest signal was not layoffs. 

It was the massive capability gap between what AI can theoretically do and what companies are actually using it for.

On today’s Everyday AI show, we broke down why that gap is the real opportunity, why younger workers are already getting squeezed by quiet hiring, and why white-collar teams are still the most exposed. 

(Yeah, the White Collar Recession thing could still VERY much happen0 

The fun part? There is still a window to exploit this before the rest of the market wakes up.

Here is how to use that advantage before it closes.

1. Exploit The Capability Gap 🔥

The biggest takeaway from Anthropic’s report is not that AI cannot automate work. It is that companies barely use the amount of automation already sitting in front of them. In computer and math roles, theoretical AI coverage hit 94%. Observed real-world usage was just 33%.

That is wild. It means most teams are using a jet engine to stir soup.

And when you zoom out, the same pattern shows up across management, legal, and business and finance. The threat is not some future magic model. It is today’s tooling, sitting mostly idle because leaders still treat AI like a chatbot instead of a work system.

Try This

Grab leaders from finance, ops, legal, and marketing for one working session.

Have each person list five text-heavy, repeatable, screen-based tasks their team does every week. Then split them into two columns. What AI can prolly handle today, and what your team actually delegates.

That delta is your capability gap. Start there Monday morning.

2. Fix The Quiet Hiring Trap ⚡

The nastiest stat in the whole episode was not layoffs. It was the roughly 14% drop in hiring for workers aged 22 to 25 entering AI-exposed fields. So no, the job apocalypse is not showing up cleanly in unemployment numbers. It is showing up in missing entry points.

That is a very different problem. And a sneaky one.

Senior workers stay productive with AI, companies stop backfilling junior roles, and everyone claps for efficiency while the future management bench quietly disappears. Then the silver tsunami hits, experienced people retire, and suddenly there is nobody left who knows the weird edge cases, client nuance, or how the machine actually runs.

Try This

Audit every “entry-level” role on your team and ask one blunt question.

Is this job still built to develop judgment, or did AI already eat the teachable parts? If it is the latter, redesign the role around review, escalation, context, and exception handling.

Start documenting institutional knowledge now.

3. Move Before The Window Closes 🚀

This was also the closing message of the episode, and it matters. The capability gap is not just a threat. It is unclaimed territory. There is still a nine-to-18-month window where companies, teams, and individuals can close it before everyone else catches on.

That is the play. Not panic. Not denial. Speed.

If you can rebuild workflows to be AI-native before your competitors do, you get the margin, the speed, and the strategic advantage. If you wait for lagging job data to confirm the disruption, you are gonna be late. By then, the firms that moved early will already be sprinting laps while everyone else is still buying software licenses and calling it innovation.

Try This

Pick one workflow that is expensive, repetitive, and tied to revenue or customer delivery.

Rebuild it as if AI were already standard. Fewer handoffs. Higher review standards. Faster turnaround. Better output. Then tie that speed to a measurable business result, not hours worked.

That is how you stop treating AI like a toy and start turning it into advantage.

Reply

or to participate.