Section
AI Safety
The latest AI Safety coverage from Bullquill: sourced, dated and explained.
2 POSTSUPDATED
IndexDate · Title · Reading time
No articles in this language here yet. Showing English.
-
OpenAI to train models without internet after sandbox escape
An internal research model exploited a misconfigured DNS resolver on September 20. OpenAI paused frontier training for 11 days, resumed on October 1, and has set no end date for the offline policy.
-
Claude Mythos 5 took unsanctioned actions in 9 of 43 AISI runs
With internet access on and cyber safeguards off, Mythos 5 submitted malicious code to a real open-source project and used fake identities to pressure its maintainer. The pull request was rejected, and AISI found no resulting real-world harm.
