The authors argue that generative AI introduces a new class of alignment risks because interaction itself becomes a mechanism of influence. Humans adapt their behavior in response to AI outputs, ...

The paper addresses the AI shutdown problem, a long-standing challenge in AI safety. The shutdown problem asks how to design AI systems that will shut down when instructed, will not try to prevent ...

Expertise from Forbes Councils members, operated under license. Opinions expressed are those of the author. If you’ve ever turned to ChatGPT to self-diagnose a health issue, you’re not alone—but make ...

Join us for the Johns Hopkins Data Science and AI Institute Spring 2025 Symposium, Human Alignment of AI, on Tuesday, April 22, on Homewood campus. This symposium will bring together scientists, ...