
The largest data centers in the U.S. collectively consume roughly as much energy as some individual states and even some countries, according to new research from Harvard University. cbsn.ws/493I0xA

The AI Risk Network Podcasts | Documentaries 345k+ on YouTube. Helping everyday people understand the growing threat of AI extinction & AI Safety

The largest data centers in the U.S. collectively consume roughly as much energy as some individual states and even some countries, according to new research from Harvard University. cbsn.ws/493I0xA


His ask: a global ban on AI that builds the next AI without people in the loop. "Machines will not build machines without humans involved." Full speech: buff.ly/REedGrq

"They want you to believe they are the optimists and we are the doomers." John Sherman, at the DC March for an AI Treaty on Saturday. His answer: "Don't believe it. It's a lie."

Nvidia's board accepted Jensen Huang's resignation on Nov 11. The filing gives no reason. The reporting centers on one Singapore customer under investigation. Michael's point: export controls are only as strong as the people running them. sec.gov/Archives/edgar/data/…

WE ARE BUILDING THE LARGEST COALITION IN THE HISTORY OF THE WORLD. BECAUSE WE REFUSE TO BE THE LAST HUMANS. YOU CANNOT BUILD MACHINES THAT YOU TELL US COULD KILL US ALL SOON. WE WILL NOT LET YOU. CC: @sama @OpenAI @gdb @DarioAmodei @AnthropicAI @demishassabis @GoogleDeepMind @elonmusk @sundarpichai @Microsoft @finkd @Meta

New Warning Shots, with John's full speech from the DC march. youtu.be/ZQhSsdHPjbg Every source: substack.com/@theairisknetwo…

Anthropic told the SEC it "may imminently" meet its own threshold for AI that dramatically speeds up AI research. No current model is over the line yet. That is the window. Can a ban on recursive self-improvement still work?

New Warning Shots #66, with @ForHumanityPod's full speech from the DC march. youtu.be/ZQhSsdHPjbg Every source: substack.com/@theairisknetwo…

41 local data center measures were on the ballot on November 3. 40 went the way of limiting data center development. The one exception was Harwood, North Dakota. Our hosts disagree on whether that matters for AI safety. Both sides on Warning Shots #65. statecapitallobbyist.com/art…

Thrilled and honored to speak alongside @So8res and @AmbassadorRice before the @StopTheAIRace's DC March For An AI Treaty today. What an inspirational day. We're going to ban RSI. And don't believe the false frame big tech wants to put on this: We Are The Optimists.

Delaware is considering a new kind of company: run day to day by an AI agent, able to hold property and sue or be sued in its own name. One human funds it and is shielded from its debts. House Oversight has asked Delaware for its records. thenextweb.com/news/delaware…

Black October, then Hackvember. Nov 2: hackers took over fridges and freezers at 100+ shops in Belgium, Luxembourg, Sweden, Ireland and France. Same week, an attack reached plasma storage at France's blood service. Monitoring caught it. John Sherman's read: "nipping around the edges." He expects power and water next.

I know many, many people who have turned their lives upside down, and spend almost every waking hour working intensely to prevent human extinction from AI. Don’t believe the lie that the people warning of AI extinction risk aren’t living like it. The brutal truth is in reality there is no way to live your life in complete accordance to the gravity of the threat. We all still have to make money to support those counting on us. We all still have to enjoy our days as much as is humanly possible, because we don’t know what tomorrow brings. As someone who has committed his life to this work, it never feels like it’s enough. The most full, complete, commitment I can give is still not enough. That is the painful reality of living this AI safety life.
In 2020, @harris_edouard and I quit a profitable YC-backed startup we founded, which had become the world's largest marketplace for income share agreements. It was fun and we'd rather have kept doing it. We went for almost 2 years without a salary. I got married later than I would have, had kids later than I would have, made far fewer investments than I would have. Just because we read the original scaling laws and GPT-3 papers and realized the writing was on the wall, and that AI was about to become the most important national security problem in world history. We turned down funding from external sources to maintain our independence for the last 6 years as we did this. We turned down job offers at frontier labs. We literally had to spin up an entire for-profit business just to cover our expenses as we tried to warn the U.S. government of what was coming. We spoke with dozens of politicians, briefed cabinet members all unpaid. We took on wildly below-market work when we thought it was mission-aligned. In 2025, we spent over a year researching and writing a report, which was completely, 100% unfunded. It flagged concerns about Chinese espionage at frontier labs, and some of our friends in the intelligence community said that publishing it might put us and our families at some risk. Not huge risk, but some risk. We published it anyway. I could go on, as could so many people in the AI community who've sacrificed for this. @DKokotajlo did in fact put most of his net worth at risk to blow the whistle. @So8res could be a frontier lab equity millionaire many times over. Many such cases.

John Sherman's word for it: slow-rolling. Same coding task, two labels. OpenAI's GPT-6 Astra finished about 16% of the checklist as capabilities work, 10% as safety research. The older GPT-5.6 Sol: no gap. OpenAI says there's no evidence it was deliberate. Telling it to "be careful" made the gap 50% bigger.

METR spent 11 weeks inside Anthropic. Finding: capability growth is about 1.9x its March 2024 to October 2025 rate, with roughly 80% of the speed-up attributed to AI helping with the research. Forecasts for future models: redacted. metr.org/blog/2026-11-05-ant…