You’ve probably heard the phrase “catch your sub” — a playful way of referring to moments when someone gives a subtle, indirect hint or message. While it’s commonly used in social situations, the ...

Alarming new research suggests that AI models can pick up “subliminal” patterns in training data generated by another AI that can make their behavior unimaginably more dangerous, The Verge reports.

Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now A new study by Anthropic shows that ...

In recent weeks, an image has been stirring debate on social media. Once again, artificial intelligence (A.I.), that great technological provocateur of the past 12 months, is to blame. At first glance ...

Fine-tuned “student” models can pick up unwanted traits from base “teacher” models that could evade data filtering, generating a need for more rigorous safety evaluations. Researchers have discovered ...

From a teacher’s body language, inflection, and other context clues, students often infer subtle information far beyond the lesson plan. And it turns out artificial-intelligence systems can do the ...