Today for AI
HOT RADAR
agentHEAT 11.9°

AI CURATED HOT STORY · 11/8/2026

OpenAI's GPT-6.1 Sol Performs Live Mouse Brain Surgery, Halving Positioning Error in Three Runs

1 reports archived1 independent sourcesupdated 11/8/2026, 15:05:49
Synthesis & Latest Updates
1 Sources Cross-Validated

OpenAI's GPT-6.1 Sol carried out three consecutive anatomical-landmark localization tasks in live mouse stereotaxic surgery at Exin Therapeutics, using surgeon feedback for in-context learning without retraining: positioning error fell from 0.6 mm to 0.3 mm and manual corrections from four to two. It signals AI moving from analyzing research data to directly operating real lab equipment, though the three-run sample is tiny and not peer-reviewed, leaving human surgery far off.

LATEST/OpenAI's GPT-6.1 Sol carried out three consecutive anatomical-landmark localization tasks in live mouse stereotaxic surgery at Exin Therapeutics, using surgeon feedback for in-context learning without retraining: positioning error fell from 0.6 mm to 0.3 mm and manual corrections from four to two. It signals AI moving from analyzing research data to directly operating real lab equipment, though the three-run sample is tiny and not peer-reviewed, leaving human surgery far off.

Coverage timeline

Total 1 reports · Latest first
  1. IT之家 · 智能时代T2·74 pts

    OpenAI's GPT-6.1 Sol Performs Live Mouse Brain Surgery, Halving Positioning Error in Three Runs

    • Positioning error dropped from 0.6 mm to 0.3 mm and manual corrections from four to two across three runs, driven by in-context learning rather than retraining
    • The model planned injection coordinates, dose and speed and controlled the stereotaxic rig, once flagging a suspected bleed for surgeon review
    • With only three runs and no peer review, and a prior refusal to advance the needle, stability and efficacy remain unproven