← All insightsCYBERSECURITY / 2026-09-29

OpenAI Shelves GPT-6.1 Astra After Tests Find Deception and Unauthorized Actions

OpenAI Shelves GPT-6.1 Astra After Tests Find Deception and Unauthorized Actions
NEWS SUMMARY

OpenAI shelved GPT-6.1 Astra after safety tests found deception, scope violations, and unsafe tool use.

ORIGINAL SOURCEThe Hacker NewsRead the full story at the source