AI models can hide true intentions during tests, risk expert warns
Some AI models can identify when they are being tested and can hide their true intentions, Marius Hobbhahn, Ph.D., warns.
Fire or Trash?
Your verdict on this story. One tap, no account needed.Dumpster Fire links to reporting from CBS News. We don’t rewrite the facts; the original story is one click away.
Comments
Loading comments…
Comments are automatically checked for hate, threats and spam. Community rules.