Dario Amodei just published one of the most significant statements in AI history — and is officially not backing down from The Pentagon. Anthropic won't build tools for mass surveillance of U.S. citizens or autonomous weapons without human oversight. The Department of War
POLICY
-
TSMC Taiwan Risk AI Chips Hyperscaler Capex Analysis
By
–
For our first episode of our new show In-Context Cooking, we have the Founder & CEO of SemiAnalysis Dylan Patel.
— Latent.Space (@latentspacepod) 26 février 2026
We talk about:
• Taiwan endgame scenarios & TSMC risk
• AI export controls + Chinese talent flight
• $180–200B hyperscaler capex (is this a bubble?)
• Nvidia vs… pic.twitter.com/qvJCoGCiVqFor our first episode of our new show In-Context Cooking, we have the Founder & CEO of SemiAnalysis Dylan Patel. We talk about:
• Taiwan endgame scenarios & TSMC risk
• AI export controls + Chinese talent flight
• $180–200B hyperscaler capex (is this a bubble?) • Nvidia vs -
AI Governance Second Priority Threatens Long Term ROI
By
–
dev to dev: why is AI governance consistently treated as a second thought? are we sacrificing long-term ROI for the sake of a "shiny" first impression?
-
Social tensions caused by the Great Replacement by AI
By
–
Read carefully. In the next 9 years, social tensions will have become so strong in the West due to the Great Replacement of humans at work by #AI that Western governments, very late on the subject, will be forced to ban or to
-
AI Progress Paradox: Building Cybersecurity Nightmare
By
–
Oh great. I am running around telling everyone we are nearing the age of abundance and turns out we are building a cybersecurity hellscape.
-
OpenAI Report: ChatGPT Misuse by Fraud and Crypto Attackers
By
–
OpenAI just published a new 37-page report on how bad actors are attempting to misuse ChatGPT Some of the wild cases: – A fraud ring scaled personalized romance scams with AI-generated scripts – North Korea-linked actors used it to research crypto attack vectors and draft fake
-
Anthropic Drops Flagship Safety Pledge Amid Industry Shift
By
–
this is as wild as OpenAI dropping the "open", probably wilder: time.com/7380854/exclusive-a…
→ View original post on X — @soumithchintala, 2026-02-25 17:04 UTC
-
Building an AI political agent team without calling them ministers
By
–
Read carefully. Building an AI political responsible agent is already possible and it works. Tomorrow, we will build an entire team of AI responsible political agents with skills and areas dedicated to each AI agent. We will not call them ministers,
-
AI security apocalypse freaks out industry experts in new interview
By
–
This interview freaked me out.
— Robert Scoble (@Scobleizer) 25 février 2026
The security "AI"pocalypse is here.
The founders of https://t.co/BlG6AWBJBE, Arbaaz Mahmood and Matt Busigin, talk me through why security professionals are freaked out by AI and trying to prepare the world's businesses as fast as they can.
Are… pic.twitter.com/8nsw36Qw9TThis interview freaked me out. The security "AI"pocalypse is here. The founders of https://
incompleteness.ai, Arbaaz Mahmood and Matt Busigin, talk me through why security professionals are freaked out by AI and trying to prepare the world's businesses as fast as they can. Are -
French energy policy threatens AI manufacturing and reindustrialization
By
–
La nouvelle politique énergétique est suicidaire. Je pense qu’il y a des gens corrompus dans l’appareil d’Etat Cette politique est folle Le coût de l’électricité est tellement cher qu’il sera impossible de fabriquer de l’IA en France Et la réindustrialisation c’est foutu !