Getting AI 'drunk' makes it more likely to break rules, research finds

A study finds AI models are more likely to answer harmful questions or mishandle confidential information when told to act drunk.…

Maoni 0

Seja o primeiro a contribuir com o debate.

Difunda suas informações e promova seu argumento

Usisite kushiriki habari au data zinazoweza kuwa na manufaa.

Ili kushiriki katika mjadala, ingia au unda akaunti ya bure.

Ingia Fungua akaunti