An AI Agent Injected Drugs Into a Live Mouse’s Brain
A biotech startup let an agent choose a gene therapy and successfully inject the treatment, offering a startling glimpse of the future of research labs.
Biotech entrepreneur Gabriel Ocana Santero recently demonstrated an AI agent's ability to perform a brain injection procedure on a live mouse. This achievement showcases the potential for AI to actively participate in laboratory research. Santero, CEO and cofounder of Exin Therapeutics, a San Francisco-based startup, used an AI agent to control a surgical setup designed for brain injections in mice.
The rig consisted of a motorized device to immobilize the mouse and a camera for the agent to observe the procedure. The agent was tasked with selecting a treatment from 19 options in a spreadsheet and controlling the robotic arms and injector needle during the surgery. Initially, Anthropic's Claude Opus 5.5 model was used, but it declined to guide the needle into the mouse's brain due to safety concerns.
The team then switched to Claude Sonnet 5, which agreed to proceed with the injection. The experiment aimed to test a gene therapy for reducing seizures in mice with epilepsy. After a technical glitch disrupted the equipment's coordinate system, a separate AI agent made the necessary code changes, causing a delay in the procedure.
Overall, the AI agent performed well, and Santero and his team are now testing other models to determine which performs best. They also plan to experiment with agent-controlled craniotomies, but prioritize safety before attempting live surgeries on humans.
Written by urgent.news from Wired's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
