Recently, neurosurgeons used an artificial intelligence system to guide them during a brain-tumor operation. This landmark first opens the door to a new practice that could become widespread, namely AI-assisted surgical interventions. The operation occurred a few months ago but was only revealed a few days ago by a major British newspaper.
A Non-Cancerous Tumor Threatening the Patient’s Vision
In 2025, the Surgical Robot Transformer-Hierarchy (SRT-H) model carried out the first autonomous surgery, namely a gallbladder removal. This AI-powered robot thus performed the first complex operation without human intervention. Before that, machines like the DaVinci system simply reproduced, with high fidelity, the exact movements dictated in real time by the hands of a surgeon behind a console.
As The Guardian reveals in an article dated 27 August 2026, human neurosurgeons used AI for the very first time in the context of a surgical operation. The intervention was carried out in May 2026 on a 48-year-old male patient suffering from a brain tumor – a non-cancerous pituitary adenoma. Measuring 11 millimeters, it was located at the pituitary gland, a small gland that is crucial because it drives most of the body’s other hormonal glands.
Close to the optic nerves and several blood vessels, the tumor was growing and threatening the patient’s vision. The operation was therefore unavoidable to prevent the patient from losing his sight. Unsurprisingly, the margin for error was razor-thin. Indeed, just a few millimeters could damage a nerve, cause vision loss, or lead to significant neurological sequelae. Thus, this challenging context prompted the neurosurgeons at the National Hospital for Neurology and Neurosurgery in London to use AI to assist them.
A Reliable ‘Digital Co-Pilot’ for a Very Delicate Operation
In practice, the professionals performed a transsphenoidal endoscopic surgery. Thus, they passed through the nasal cavities with a fine camera to reach the base of the skull. Developed by University College London, the AI acted as a “digital co-pilot,” analyzing the camera’s video feed and live-highlighting the anatomical structures to avoid. Notably, the powerful device involved was NVIDIA’s Clara IGX platform. The aim was to run the AI locally to avoid the disastrous consequences of a potential Wi‑Fi cut at the hospital.
“Thanks to the analysis of hundreds of surgical videos, the system has been exposed to a wide variety of surgical examples that a surgeon would take years to encounter. It is designed to help recognize in real time the critical anatomy, the surgical instruments, and tissue interactions, thus assisting the surgeon during extremely delicate interventions,” the article states.
Originally, the patient had been treated in 2024 after a severe epilepsy episode. Doctors then diagnosed a tumor pressing on his optic nerves, gradually erasing his peripheral vision and affecting his ability to walk. The operation was a success, allowing the tumor to be removed safely. Less than a week after the procedure, the man walked independently and had fully regained his sight.
