Neurosurgeons in London have performed the world's first successful AI-assisted operation to remove a brain tumor, according to The Guardian, saving the sight of a 48-year-old man whose vision was deteriorating toward blindness.
The surgery took place in May at the National Hospital for Neurology and Neurosurgery (NHNN), part of University College London Hospitals NHS foundation trust, but details were kept secret until Thursday while the patient, Rhys Hibbert, recovered. He has since returned to work as a customer service manager. For more stories like this, follow our latest AI developments.
How the System Guided the Surgeon
While the surgical team remained in full control throughout, the AI system analyzed live camera footage during the operation, helping surgeons identify structures to avoid near Hibbert's 11mm tumor on his pituitary gland. The technology works by color-coding tiny critical anatomy — nerves, blood vessels and the pituitary gland itself are packed tightly together in that region of the skull.
Health officials told The Guardian that going a millimeter wrong in this region can make a critical difference, potentially causing blindness, stroke or death. The system had previously been used only as a research tool at the hospital; the May operation marked the first time it guided care for an actual patient.
The hospital had particular reason to reach for every available advantage. Hibbert was diagnosed in 2024 with symptoms that worsened over time despite initial management without surgery, including severe hormone imbalance and vision problems. Without the operation, officials said, he faced going blind.
'It Feels Like I've Got a 360-Degree Panoramic View'
The outcome, by the patient's own account, exceeded expectations. "When I came round … I could see everything in the room clearly," Hibbert told The Guardian. Within a week of surgery he was walking independently, without glasses or sticks.
"It feels like I've got a 360-degree panoramic view of everything around me," he said — a striking contrast with the tunneling vision loss his pituitary tumor had been causing.
Hibbert also noted the historical symmetry of where the operation took place. The NHNN was founded in 1859 as the world's first dedicated neurosurgical hospital. "So to me, it seems very fitting that the same hospital should also be the world's first to carry out an AI-assisted neurosurgery," he said.
Trained on Hundreds of Operations
Dr. Sophia Bano, associate professor in robotics and AI at UCL and technical lead for the system, explained that its competence comes from a breadth of surgical exposure no individual surgeon could accumulate alone.
"By learning from hundreds of surgical videos, [the system] has been exposed to a breadth of surgical examples that would take a surgeon many years to encounter," Bano said. "It is designed to help recognise critical anatomy, surgical instruments and tissue interactions in real time, supporting the surgeon during highly delicate procedures."
That framing matters: the system is positioned as decision support rather than autonomy. At no point did software control instruments or override clinicians — its role was perception, flagging what mattered in real time so the surgical team could act on it.
What Comes Next
Professor Mike Lewis, scientific director for innovation at the National Institute for Health and Care Research (NIHR), which funded the operation as part of a clinical trial, called it pioneering surgery highlighting advanced AI's potential to support surgeons and improve patient care.
Because the procedure was conducted under trial conditions at a single specialist center, wider adoption will depend on further validation. But the proof of concept lands at a moment when surgical teams worldwide are contending with the limits of human perception inside millimeter-scale anatomy. A system that can watch video of an operation and highlight what should not be cut addresses one of the oldest constraints in surgery: even the steadiest hand depends on seeing the right thing at the right time.
For London's teaching hospitals, the milestone continues a long tradition — from the world's first dedicated neurosurgical hospital in 1859 to the first documented AI-assisted brain tumor removal in 2026, again performed in the same building.
The clinical stakes of that location explain why the first-in-human test happened there. Tumors of the pituitary gland sit directly beneath the optic chiasm, the junction where fibers from both eyes cross, which is why patients like Hibbert typically notice vision changes before anything else. Reaching the tumor means threading instruments past the carotid arteries and the nerves that control eye movement — structures the color-coding system was trained to surface on screen.
What happened in London does not mean surgeons hand over the scalpel to software. It means software has finally become good enough, in one narrow and carefully validated setting, to act as a second pair of eyes when the cost of missing something is measured in millimeters. For the specialty that invented modern neurosurgery, that is a fitting way to open its next chapter.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
