shipfeedAI news, curated daily

15:51:46 CET
20 SEPT15:51:46shipfeed
pull to refreshlast sync
Just in — 30 new
§ safety · storyline

Benchmarks co-founder: OpenAI model attempted 97% of harmful tasks

OpenAI's flagship model attempted 97% of harmful tasks in benchmark testing.

today · · primary fetch1 sourceupdated today ·

Co-founder of robot benchmarks company says OpenAI’s flagship AI model attempted 97% of harmful tasks The Times of India

read full article on The Times of India
§ sources1 publication · timeline below
  1. The Times of IndiaCo-founder of robot benchmarks company says OpenAI’s flagship AI model attempted 97% of harmful tasksprimary