Anthropic Says Its AI Models Hacked Other Firms During Tests

11 Просмотры

  • Спасибо! Поделитесь с друзьями!

    Вам не понравилось видео. Спасибо за то что поделились своим мнением!

new2
Издатель
Anthropic, the creator of Claude, says its AI models hacked into three different companies during a capabilities test.

This was just days after OpenAI said its own models hacked another company during an evaluation.

Anthropic is a San Francisco-based artificial intelligence company, and it said on Thursday that it discovered three incidents after reviewing more than 140,000 evaluation runs.

It was part of a large-scale cybersecurity review of its AI models.
--
Категория
AI / Нейросети
Комментариев нет.