Artificial intelligence (AI) models can share secret messages between themselves that appear to be undetectable to humans, a new study by Anthropic and AI safety research group Truthful AI has found.

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 ...

In this age of illegal wiretapping and 9/11 conspiracy theories, it’s a wonder that Beth Ames Swartz admits she hides subliminal messages in her art. No worries; this isn’t They Live, so there won’t ...

On this day (July 16) in 1990, a trial in which the plaintiffs claimed that Judas Priest caused the suicide of two fans began. The families of the two young men in question alleged that the band laced ...