Sam Altman told OpenAI employees there are three ways the race to safe AI could end, and that he favors the one modeled on nuclear inspections. The Information reported it on October 19, 2026, in a story headlined “Altman Describes Three Endgames for AI, Including One Where a Single Firm ‘Melts Everyone Else’s GPUs’“, by Stephanie Palazzolo and Amir Efrati. Efrati’s own summary on X lists the three: an IAEA for AI, a CERN for AI built by merging the labs, and what Altman called the Eliezer plan.

— The IAEA Plan: An international agency enforces safety standards on frontier labs
— The Demis Plan: Labs merge into, or work through, one organization that supplies models to cooperating countries, a CERN for AI
— The Eliezer Plan: One firm reaches recursive self-improvement and uses the power to melt every rival’s GPUs

— Read More

Here’s a speech I gave on Saturday the 17th of October at the IMF meeting in Bangkok. This speech is inspired by the megatrend readers have seen me cover this year of AI beginning to operate at the frontier of human science, sometimes autonomously. While we all may quibble over the details, I think it’s evident we’re in the early stages of the singularity, and I wanted to use my privileged position of giving the Per Jacobsson lecture to try and tell the assembled audience of economists, bankers, and policymakers, that they should expect significant, rapid change in the future. In this speech, I focused primarily on how humanity can share in the benefits of AI if we exercise proper foresight and restraint, and I’m going to be expanding on the stakes of the singularity and the risks that we’ll all face in coming issues.

As I said a few times in the speech – some of which was ad libbed and not captured here – I take my own responsibility here extremely seriously. I find myself both a narrator of what is happening in AI, a principle agent behind what is happening in AI, and a translator that goes between many communities. In all of this, I am seeking to be truthful and to be clear about my own culpability. This is a forever work-in-progress. — Read More

What I’m about to tell you has, I realize, become something of a cliché: I resigned this week from OpenAI. I led the writing of the safety reports we published with each major launch. Now I’m joining a parade of former colleagues—at OpenAI and the industry’s other leaders—who have decided that the current path is unacceptable.

I agree with other recently departed staff that the companies building this technology aren’t being nearly careful enough. But I believe that we need to look deeper than specific rules or new laws. We need to talk about culture. — Read More

Over two years ago we wrote a detailed deconstruction of p(doom) and argued that its primary effect is to launder vague intuitions and fears through a facade of quantification. The essay has fresh relevance today as p(doom) rhetoric is driving public discourse and policy attention to an unprecedented degree. Unsurprisingly, today’s probability estimates of AI existential risk are no more rigorous than the ones from 2024. Hence this repost. — Read More

#singularity

There has been, over the last week, what I think is a healthy debate about the philosophy and psychology that undergirds the views of meaningful segments of the AI community, particularly those obsessed with doomsday scenarios. It is, in the end, difficult to reason with a philosophy that grants equivalent moral weight to not just all beings — human or not — who exist today, but who may ever exist in the future; this tilts the scales in such an absurd fashion towards safetyism that innovation is impossible and freedom is intolerable.

Worse, it taps into the psychology of religion, where dissent is not brooked and questioning the premise is heresy. I guess that makes me a heretic then: I reject the premise in favor of doubt in our ability to foresee the future, combined with faith in humanity figuring things out along the way. — Read More

Nvidia chief executive Jensen Huang has described warnings that AI could lead to humanity’s extinction by the next decade as “doomsday narratives”.

“2030 is not going to be the end of the world. There is 0% chance that’s going to be the end of the world”, Huang said in an interview with CBS News, the BBC’s US partner.

Huang added: “Scaring people is unnecessary. It is irresponsible.” — Read More

AI getting smarter does not automatically make it uncontrollable. The real question is what we give that intelligence access to. Give a model enough permissions, persistence, resources, copies and coordination, and its mistakes can compound into real-world power. The good news? Those layers are ours to design.

… Most people assume that intelligence, pushed far enough, naturally turns into uncontrollability, and that the smarter the mind, the less any fence can hold it.

I think that assumption is wrong. I don’t think AI is on an inevitable path to killing us all.

My rationale has nothing to do with the technology being harmless. My reason is that we have misunderstood where an AI system’s power comes from and what we can do about it — Read More

At first, nothing seemed unusual. The frontier laboratories were still training larger systems. Chip orders were still rising. Data center campuses were still expanding. New reasoning models were still arriving with better scores in coding, science and mathematics. From the outside, the AI race appeared to be accelerating exactly as expected.

Yet something had changed in the documents surrounding those models. New capability thresholds appeared. Security requirements became stricter. Risk reports grew more candid. Safety cases moved closer to launch decisions. Some internal activities were paused until stronger controls were installed. There was no dramatic announcement that the race had stopped. Instead, the brake emerged as a collection of technical clauses, governance gates and operational restrictions.

That is what makes the moment fascinating. The frontier has not run out of ambition. It has begun to encounter a harder problem: capability can now move faster than the mechanisms used to measure, contain and approve it. The question is… — Read More

Anthropic just put numbers [in June] on recursive self-improvement, their AI accelerating AI. The stats are staggering. The questions they raise are bigger.

Right now, more than 80 percent of the code merged inside the company that makes Claude is written by Claude. Not by its engineers. By the model they built.

A year ago that number was in the single digits.

Anthropic put this in writing, and the phrase they reach for is one that has haunted AI circles for years: recursive self-improvement.  — Read More

Mankind has spent 5,000 years trying to create more intelligence.

Now we may have figured out how to manufacture intelligence itself — and right on schedule, somebody wants to regulate the living Hell out of it.

…Welcome to America, baby.

Miracles happen, tech miracles, steward them thoughtfully. Don’t strangle them. — Read More