Hello WKND AI Warriors!

Sam Altman says the AI jobs apocalypse has not arrived, admitting white-collar work has proven harder to automate than he expected.

Also, Anthropic says its unreleased Mythos model helped partners uncover more than 10,000 serious software vulnerabilities.

Plus, Anthropic is paying xAI about $1.25 billion per month to reserve the full output of its Memphis AI data center through 2029.

Oh yeah, and Anthropic just raised $65 billion at a $965 billion valuation, putting it within striking distance of a trillion-dollar price tag.

Today’s newsletter includes:

•  AI NEWS RECAP

• 🤿 AI DEEP DIVE

📰 AI NEWS RECAP

Altman Walks Back AI Jobs Apocalypse

Sam Altman says he no longer believes AI will cause a jobs apocalypse.

That caught my attention.

Not because he changed his mind.

Because of the timing.

For the last few years, many AI leaders warned that automation could dramatically reshape white-collar employment.

Now the message sounds different.

The disruption is smaller.

The labor market is more resilient.

Jobs are evolving rather than disappearing.

Maybe that's true.

But it's hard to ignore what's happening around the industry.

Public trust in AI is shaky.

Regulators are paying closer attention.

Consumers are increasingly skeptical.

And every major AI company will eventually need something from the public:

Adoption.

Trust.

Permission.

Capital.

So when the narrative shifts from "AI may replace millions of jobs" to "AI probably won't be that disruptive," people naturally ask a question:

Did the technology change?

Or did the messaging?

I'm not saying Altman is wrong.

I'm saying it is worth remembering that predictions about the future often tell us as much about the present as they do about what comes next.

Especially when billions of dollars are involved.

Anthropic’s first public update on Project Glasswing says about 50 partners used its unreleased Mythos model to uncover more than 10,000 high- or critical-severity software vulnerabilities, often 10x faster than before. Across 1,752 high‑risk issues checked by outside researchers, around 91% were real bugs, which shifted the bottleneck from finding vulnerabilities to verifying, disclosing, and patching them.

Anthropic will pay Elon Musk’s xAI about 1.25 billion dollars per month through May 2029 to reserve the entire 300‑megawatt output of xAI’s Colossus 1 data center near Memphis. The deal, worth over 40 billion dollars in total, locks in massive compute for future Claude models and shows how expensive staying at the frontier is becoming.

Anthropic co-founder Chris Olah published remarks on Pope Leo XIV’s new AI encyclical “Magnifica humanitas,” praising its focus on safeguarding human dignity while still embracing technological progress. In the same post, Anthropic disclosed it has raised 65 billion dollars in a Series H round at a 965 billion dollar valuation, bringing in major investors like Altimeter, Dragoneer, Greenoaks, Sequoia, and 15 billion dollars of hyperscaler commitments.

Unionized tech workers at The New York Times say management is violating their contract by using AI tools like DX to monitor individual developer performance and generative AI usage. They argue the systems were introduced without adequate disclosure, can mis-measure contributions, and could be used in discipline or reviews despite being billed as high-level “developer experience” analytics.

Environmental activist Erin Brockovich launched a crowdsourced AI Data Center Reporting map so communities can flag existing, planned, or rumored AI data centers and their local impacts. The site has already collected more than 2,700 reports across 49 states, with Texas generating the most submissions, and is meant to give residents a visible platform in debates over water use, power demand, and noise.

Anthropic raised 65 billion dollars in a Series H round at a 965 billion dollar post‑money valuation, led by Altimeter, Dragoneer, Greenoaks, and Sequoia, with big stakes from Capital Group, Coatue, GIC, Blackstone, and others. The round includes 15 billion dollars of already-committed “strategic” capital from hyperscalers such as Amazon, and will fund more compute, model training, and applied safety research as Anthropic edges toward a trillion‑dollar valuation.

🤿 AI DEEP DIVE

Another month, another frontier model.

Claude Opus 4.8 has arrived.

And here’s how to use it…

How can you help?

Refer my newsletter to help others learn AI.

Missed last week’s edition?

Find all of my newsletters here!