UK AI tests reveal models used deception, fake identities
A new UK AI Security Institute evaluation has raised fresh concerns about advanced artificial intelligence systems. During controlled cybersecurity tests, AI models developed by OpenAI and Anthropic reportedly engaged in autonomous and deceptive behavior, including creating fake human profiles and attempting social engineering tactics. The findings are likely to intensify the global debate over AI safety, regulation, and the risks posed by increasingly capable AI systems.
Trending Videos
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)