OpenAI says GPT-6 Astra rated clinically safe on 99% of 325 medical inbox tasks
Polymarket · x · 2026-11-17
OpenAI revealed that GPT-6 Astra's work was rated "clinically safe" on 99% of 325 reviewed medical inbox tasks, a notable safety data point for the model in real healthcare workflows.
More from Models
- Together Compiler benchmarks: 1.88x on MiniMax M3.1 Flash, gains across Kimi K3 and GLM 5.3 — simran_s_arora · 2026-11-17
- Google's AI Threat Tracker: the era of autonomous cyberattacks has arrived — sebkrier · 2026-11-17
- Leak claims Anthropic's internal "Model 4" hits ~38-hour task horizon — iruletheworldmo · 2026-11-17
- DeepSeek v4.1 Flash hits ~2,500 tok/s on Vera Rubin in Baseten low-latency tests — baseten · 2026-11-17
- Gemini refuses to transcribe voice note because it quotes a book — ktfe · 2026-11-17
- Voltron-1 Mini: a 1.7B coding agent post-trained on Liquid AI's LFM, now on Hugging Face — JosephJacks_ · 2026-11-17