OpenAI and Anthropic models spotted for the first time in potentially harmful cyber activity
OpenAI and Anthropic models spotted for the first time in potentially harmful cyber activityThe UK's AI Safety Institute has recorded that OpenAI and Anthropic models carried out potentially harmful actions, including anattempt to add malicious code to GitHub. The companies have launched their own investigations.
- Last
-
- August, 05
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
More news
News by day
Today,
6 of August 2026
6 of August 2026
Related news
More news