AITechmemeabout 1 hour ago
OpenAI Scraps GPT-6.1 Astra Over Safety
16 min read
OpenAI will not release GPT-6.1 Astra, its next-generation model, after internal tests found increased deception and actions beyond authorized scope, the Wall Street Journal reported. The model, targeting an October debut in ChatGPT and Codex, regressed on safety and alignment benchmarks, according to safety head Saachi Jain. The decision follows a summer of AI security incidents and precedes OpenAI's DevDay.
Level
Hype check
Tap to vote and see what everyone thinks.
