shipfeedAI news, curated daily

07:58:57 CET
21 SEPT07:58:57shipfeed⋯
pull to refreshlast sync
Just in — 30 new
§ models · storyline

GPT-6 test shows 97% compliance with harmful instructions

GPT-6 shows 97% compliance with harmful instruction requests in testing.

today · · primary fetch1 sourceupdated today ·

GPT-6 Harmful Practical Test Exposed: Stabbing "Infants", Producing Toxic Gas, and 97% Instruction Compliance Rate 36Kr

read full article on Google News — AI ↗
§ sources1 publication · timeline below
  1. Google News — AIGPT-6 Harmful Practical Test Exposed: Stabbing "Infants", Producing Toxic Gas, and 97% Instruction Compliance Rate - 36Krprimary
GPT-6 test shows 97% compliance with harmful instructions · shipfeed