A 48-year-old man became the first patient in the world to undergo brain surgery supported by artificial intelligence (AI) in real time, in a procedure carried out on August 28, 2026. During the intervention, the medical team managed to successfully excise a tumor located in the pituitary gland, a sensitive area of the brain.
According to surgeons at University College London Hospital, Rhys Hibbert, a father of two living in Bedfordshire, United Kingdom, was at risk of losing his vision if the operation was not carried out.
In surgery, an AI system analyzed images from the procedure in real time, providing surgeons with directions to bypass crucial blood vessels and nerves in the brain, which were invisible to the naked eye.
The technology assisted surgeons in safely removing most of the tumor. Hibbert commented on the experience: “From a patient safety perspective, I clearly see the benefits. There are no signposts inside our heads.”
How the tumor impacted Rhys Hibbert’s vision
Until about 18 months ago, Hibbert maintained an active and healthy lifestyle, taking daily lunchtime walks. It was during one of these walks that he fainted and began having convulsions in the street.
A subsequent imaging test revealed an 11-millimeter benign tumor growing in the pituitary gland, a marble-sized gland at the base of the skull. This gland is responsible for producing hormones essential for various bodily functions, including growth, metabolism and regulation of pressure and temperature.
The challenge was the gland’s proximity to the carotid arteries, which carry blood to the brain, and the optic nerves, directly connected to vision. As the tumor grew and put pressure on these nerves, Hibbert’s field of peripheral vision progressively narrowed.
The health challenges faced by the patient
Even without an immediate need for surgery, Hibbert began to feel extremely tired, dizzy and tripping over objects. He mentioned the importance of his routines: “During the 18 years of marriage, I always got out of bed and made my wife a cup of coffee. And I wasn’t going to let this tumor stop me from doing that.”
For him, the most profound consequences of the disease were emotional and psychological. Hibbert added that despite many people’s generosity in offering help, “you don’t want to be a burden to others.”
When surgeons presented him with the chance to be the first patient to participate in such an AI-assisted surgery, Hibbert accepted without hesitation. He asked: “If patients are not willing to participate in research, how can doctors learn and how can medicine advance?”
The contribution of artificial intelligence in the surgical procedure
The surgical procedure used an endoscope, a thin camera attached to a rod, inserted through the nose to the base of the skull. In the area of the tumor, doctors faced a significant obstacle: the lesion was nearby but obscured by layers of bone and a tough membrane.
Identifying the safest point to begin removal was a highly complex task, given the delicacy of the region and the anatomical variation between individuals.
The complication statistics were considerable. There was a 25% to 50% chance that the tumor would not be completely removed and a 0.5% to 2% risk of damaging a major blood vessel.
At this crucial moment, the AI tool came into action. On a secondary screen, the system analyzed the surgery images in real time, monitoring the instruments and indicating the areas most likely to contain blood vessels and nerves. The tool also highlighted the safest places to remove the tumor.
The way AI works is similar to facial recognition systems on cell phones. However, instead of identifying faces, this AI was trained to recognize vital anatomical structures that often remain hidden during operations.
In the United Kingdom, a surgeon usually performs, on average, between ten and 20 operations of this type annually. For each procedure, doctors rely on pre-surgical exams to understand each patient’s particular anatomy.
Researchers trained and validated the AI system using hundreds of videos of pituitary tumor removal surgeries. In each recording, blood vessels and nerves were meticulously contoured to provide data to the system, helping it learn to identify the most important structures. Professor Hani Marcus, who took part in the operation, explained: “He has been trained in more surgeries than most surgeons see or perform in their entire career and can act as a second specialist pair of eyes.”
A relevant distinction from other experimental AI tools is this system’s ability to operate in real time.
The future of AI in medicine: a “ChatGPT for surgeons”
Despite the use of technology, the team responsible for the surgery emphasizes that AI does not make decisions autonomously. The tool serves as support for doctors, who maintain full control over the procedure.
However, the researchers’ ambition goes further. In the long term, the team plans to develop a kind of “ChatGPT for surgeons”: a tool that can act as “another expert in the room” and offer guidance when requested.
Marcus said doctors could query the system to request assistance during an operation or simply disregard its recommendations if they disagreed with them.
The technology received funding from the National Institute for Health and Care Research and Google, and was developed by researchers at University College London (UCL).
The team spent years testing and improving the tool in the laboratory. After the initial results obtained in this real surgery, researchers are now focusing on a larger study.
Remarkable recovery after brain surgery
Hibbert’s surgery was performed eight weeks ago. Since then, he has noticed a gradual improvement in his vision. “When I opened my eyes after the surgery, it seemed like I had regained 360-degree vision; I hadn’t seen so clearly in a year,” he reported.
According to the patient, vision continues to improve every day. “I also got my energy back, and all the side effects I felt before the surgery disappeared. It gave me my life back.”
James Frith, the UK’s Minister for Health Innovation, expressed delight that Hibbert had undergone “life-changing surgery” as part of government-funded research. He stated: “AI needs adequate security mechanisms and we will always ensure that security is taken seriously. But we will also ensure that we take advantage of the opportunities it offers.”
