How to keep dangerous AI out of reach while preserving model choice

You don’t need to believe everyone dies. Staying in control requires understanding, coordination, and foresight. Eight failed arguments, why they fail, what we can do together, and why.

An incentive program for security work, and cross-over lessons from OpenAI’s Hugging Face incident

The protections keeping capable AI away from malicious intent are weakening. What developers and software companies should be doing now, with old techniques and new ones

The protections keeping capable AI away from malicious intent are weakening. What developers and software companies should be doing now, with old techniques and new ones

Two labs skipped monitoring on their cyber evals. The same week, a hundred companies signed a letter supporting deployments that can’t be monitored.

OpenAI accidentally hacked Hugging Face. A basic layer of defense appears to have been missing, and no one has explained why.

A better definition of slop points to feed control as the priority.

Rethink Medical Backlash Against Pervasive Testing

Opinions about AI are outrunning our understanding of the software beneath it

A layer-by-layer look at the AI value chain — chips to applications — finds firmer footing on capex, but a mountain only half climbed.

The wrong question — and a better one for reading the AI boom.

The wrong question — and a better one for reading the AI boom.

Blaming AI for low-quality content misses the real problem—and the real opportunity.

Blaming AI for low-quality content misses the real problem—and the real opportunity.

One-sided fixes become weapons. The exit is collaborative.

Lazy Mistakes in Hiring and the Truth Behind Jobs Data

Lazy Mistakes in Hiring and the Truth Behind Jobs Data