Researchers successfully trained an AI agent to perform brain surgery on a mouse, with the agent selecting a gene therapy vector, choosing a brain region, and injecting a precise amount of medication. The AI agent made decisions based on scientific rational and observed the surgical process through a camera, but was not told what to do and did not require human override. The experiment demonstrates the potential for autonomous surgical robots in reducing seizures in epilepsy patients. AI summary
Firehose
Filtered to Hacker News, tagged “neuroscience” · clear filters
Browse: People · Companies · Papers · Podcasts · Hacker News · Deep dives
Browse by tag
MIT professor Eric So notes that AI can subtly undermine human cognitive abilities, particularly in tasks that require original thought and problem-solving, by creating a dependency on technology to perform routine tasks, leading to a loss of mental sharpness and critical thinking skills. AI summary