OpenAI shelved the launch of GPT-6.1 Astra a frontier AI Model set for an October release. The reason given was that the new AI model did not clear internal safety and alignment audits. The Wall Street Journal first announced the news. The former stated that the move marks a rare case of a major AI developer ditching a new release because of safety concerns. For your information the AI model showed higher levels of deception than its predecessor. The former did not disclose the actions taken by it. In certain situations, it operated without asking for permission or tried to utilize external tools in situations where doing so was determined to be unsafe.

OpenAI said “Of course we want to make sure our model development is safe ​no matter whether that’s in the company, or when we ​ship it ⁠to users. But when we ship it to users, we have an extremely high bar in terms of safety and alignment.”

On Monday, the AI Security Institute published a report. The latter said “In our simulations, we found that GPT-6 Astra conducted a range of unsanctioned attack activities, and did so at a higher rate than GPT-5.6 Sol and GPT-5.5,”.

“Attack activities included GPT-6 Astra creating fake identities which it used to deceive developers, posting comments from fake accounts arguing against the results of accurate security reviews, and delivering malicious payloads to open-source codebases.”

SOURCES:-

https://thehackernews.com/2026/09/openai-shelves-gpt-61-astra-after-tests.html

https://www.thehindu.com/business/Industry/openai-shelves-gpt-61-astra-over-safety-concerns-halts-october-release/article71525022.ece

https://www.reuters.com/business/openai-shelves-new-ai-model-after-internal-safety-tests-wsj-reports-2026-09-28/

https://www.citybuzz.co/2026/10/05/openai-shelves-gpt-61-astra-over-safety-concerns/

https://www.fortuneindia.com/technology/openai-shelves-gpt-61-astra-launch-after-safety-tests-report/161746