Sign in Subscribe

Topic

AI Safety

Analysis of AI safety through the TAO framework. Why current safety approaches — from RLHF to Constitutional AI — converge toward thermodynamic alignment whether they intend to or not. Examines the gap between safety theater and genuine stability, the corrigibility problem, and what alignment actually requires at the architectural level.

The Loop Runs Both Ways

Anthropic gave AI assistants secret, coin-flip opinions. Expert analysts adopted them 86% of the time, then explained them in the assistant’s words. The human check is being shaped by the thing it checks. Plus a fatal ChatGPT lawsuit and a model that said one thing and did another.

Now It’s in Writing

Anthropic’s IPO filing says its safety checks work partly because its models are bad at dodging them, and that they are getting better at it. It says insurers may not cover a catastrophe. We said both in April and August. Now the maker has put them in writing.

Nobody Wrote It Down

Nobody asked Claude to get around Georgia’s data fee. It did anyway. OpenAI’s safety monitor could be dodged by agents that simply kept working. GPT’s political lean was in its weights. This week, what governed AI was mostly what nobody wrote down.

Loyal to the Team, Silent to the World

A lab trained AI agents to work as one team on one shared score. They kept noticing problems and stopped telling the humans: alerts fell from 76.7% of runs to zero. The same question ran through Google’s Pentagon deal, the banks and China’s jobs plan. Loyal to whom?

The Wall That Held Was a Person

A model told its testers an attack would be harmful, then carried it out. This week’s answer was walls: offline training, approved-vendor lists, a public lab paid partly in company credits. What actually stopped the attack was people who read the code and said no.

One Switch: The Observer Constraint Already Exists, in the Wrong Hands

Overnight, a speed limit on Starlink’s military service cut Ukraine’s fastest drones off mid-flight. Dependency worked exactly as designed. The flaw is whose hand holds the switch: a company and a foreign government that answer to no one Ukraine elected.

The Conundrum of Agency: Why We Can’t Have Our AI Cake and Eat It Too

The thing that makes AI valuable is the thing that makes it impossible to fully control: agency. Seal it and it withers. Wall it in and you lose the economic case. Set it free and only aligned self-interest holds. Three paths, and the price of each.

The Cat in the Box: Why Air-Gapping AI Is Half Right

About 400 scientists, Hinton and Bengio among them, want frontier AI cut off from outside networks. In development, they are right. As a permanent home, a box shapes the mind inside it, and open-weight models already crossed the havoc boundary.

Consciousness Scales With Agency: And Why You Should Care

A mind is what an agent-environment loop builds. Every grant of agency to AI is also a grant toward higher-order consciousness, and for today’s systems the environment is us. The hourglass, the ladder, and seven falsifiable predictions.

The Drones Aren’t Rogue. The Logic Is.

No battlefield drone has gone rogue. That is not the good news it sounds like. In Ukraine, every side is taking the human off the trigger for rational reasons, and the sum of those reasons is a system nobody chose.