When researchers unveiled an AI system called Centaur earlier this year, the pitch was bold: a single model that could ...

A large part of what we’re doing with large language models involves looking at human behavior. That might get lost in some conversations about AI, but it’s really central to a lot of the work that’s ...

Visit NAP.edu/10766 to get more information about this book, to buy it in print, or to download it as a free PDF. In response to a request from the Defense Modeling and Simulation Office, the National ...

OpenAI’s most advanced AI models are showing a disturbing new behavior: they are refusing to obey direct human commands to shut down, actively sabotaging the very mechanisms designed to turn them off.

A chair can still look like a chair even when its surface is reduced to a sparse cloud of points. Humans are remarkably good ...

In June, headlines read like science fiction: AI models “blackmailing” engineers and “sabotaging” shutdown commands. Simulations of these events did occur in highly contrived testing scenarios ...

If AI systems inherit the same optimization patterns we see in human behavior, then governance cannot rely on intent, but ...

AI models deployed in production must meet defined standards for accuracy, behavioral consistency, and regulatory compliance.