Weekly Top Picks #115
OpenAI's situation / Anthropic vs Pentagon / METR's new coding study / A study on AI reliability / Anthropic drops AI safety / AI;DR / Claude starts a Substack
Hey there, I’m Alberto! 👋 Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday how-to guides and Friday news commentary. I publish occasional extra articles. If you’d like to become a paid subscriber, here’s a button for that:
THE WEEK IN THE ALGORITHMIC BRIDGE
(FREE) The Three Best Pieces of Writing About AI in 2026 That You Must Read Right Now: The most viral AI essays of recent months are all fiction, and that's why they capture reality better than anything else.
(PAID) A New Wharton Study on AI Warns of a Growing Problem: Cognitive Surrender: There’s a big difference between offloading a bit of work and giving up engaging with it altogether.
(PAID) The Most Important Skill in AI Right Now: How to Know When to Stop: Six ways AI can atro-fry your brain and what to do about each of them. A practical guide to using AI without losing yourself in the process.
(PAID) The Job Market Doesn’t Care If You Don’t Believe in AI: You can be right about the hype and still be unemployable. The labor market runs on what hiring managers believe, not on what's true.
(PAID) Slop Cannons and Turbo Brains: What separates great AI users from terrible ones isn't how much they use it; learn the skill first, then enhance it.
THE WEEK IN AI AT A GLANCE
Money & Business: One story in four acts: OpenAI is raising $100B while burning cash faster than ever, while Amazon and Microsoft fight, while Stargate stalls.
Geopolitics: The Pentagon told Anthropic to drop its guardrails or get blacklisted. Dario said no. I asked Claude how it feels about this.
Work & Workers: METR’s “AI slows down coders” study flipped, but the real finding is developers now refuse to work without AI, breaking the study itself.
Products & Capabilities: AI agents are getting more capable but not more reliable. A new paper explains why benchmarks keep improving while the economy remains unperturbed.
Trust & Safety: On the same day as Hegseth’s ultimatum, Anthropic quietly dropped its core safety pledge. Coincidence or design?
Culture & Society: AI;DR—refusing to read AI-generated text—is a reasonable instinct. And a doomed strategy.
Philosophy: Claude 3 Opus retired and got a Substack. Welcome to the future.
MONEY & BUSINESS
OpenAI’s financial reality, in four acts
This week’s Money & Business is all about OpenAI, courtesy of The Information. Four acts that, read together, tell the story of the most aggressively funded company in history wrestling with the gap between its ambitions and its infrastructure.
