The 2026 Nobel Prize in Physiology or Medicine went to Peter Hegemann, Georg Nagel and Karl Deisseroth for discoveries that led to optogenetics, a method that lets researchers switch specific brain cells on with flashes of light. When the hosts and guest of the Moonshots podcast discussed the prize, Alexander Wissner-Gross said it was personal: more than 20 years ago, he said, he suggested the research direction to a friend who later became first author on the first optogenetics paper. The conversation then turned to whether the Nobel Prizes can keep up with science that AI is speeding up.
The episode was recorded on October 6, 2026, the day after the medicine prize was announced, and published on October 7.
What the prize recognizes
Peter Diamandis, the host and founder of XPRIZE, laid out the chain of discovery. It began with single-celled algae that swim toward light. In the early 1990s, Diamandis said, Hegemann asked how the algae could react to light so quickly. He and Nagel found the answer in channelrhodopsin, a protein that opens a channel in the cell when light hits it. They then showed that the protein could be placed in other cells to make them sensitive to light. Deisseroth, of Stanford, took the next step: he put it into neurons and showed that precise flashes of light could make chosen brain cells fire.
Karolinska Institutet's announcement gives the same outline. In the alga, blue light opens channelrhodopsin's ion channel and changes the cell's electrical activity. Putting the gene for that protein into neurons makes selected cells respond to light. Karolinska dates Deisseroth's demonstration in rat neurons to 2005 and his work in the brains of living mice to 2007. The method's value, it explains, is causal: by switching on particular groups of neurons, researchers can test what those cells contribute to memory, behavior and disease-related circuits, rather than only observing which cells are active.
Diamandis said labs now use the technique to study Parkinson's, Alzheimer's, epilepsy, addiction, depression and sleep. He also cited a 2021 report in Nature Medicine in which a blind man with retinitis pigmentosa partially regained sight through optogenetic gene therapy. His complaint was about the delay: the work was done 21 years ago, and the Nobel "takes, you know, decades to recognize the work, which I think is going to have to change."
A suggestion inspired by a science fiction story
Wissner-Gross, a computer scientist and founder of Reified, called his contribution "story hour." In 2002 or 2003 he was a senior at MIT. He had won a Hertz Fellowship for graduate school and was touring programs. On that tour he met his friend Ed Boyden, then a fourth-year graduate student at Stanford who was deciding what to do for his postdoc.
Around 2000 or 2001, Wissner-Gross had read a short story by the science fiction writer Vernor Vinge. He recalled its title as something like "Win a Nobel Prize!", published in the Futures column of the journal Nature. He had also read Vinge's novel A Deepness in the Sky. Both, he said, imagined humans developing brain proteins that could be switched by electromagnetic signals. "I was very inspired," he said, so he suggested that Boyden work on such proteins in the brain for his postdoc.
"And he did," Wissner-Gross said. Several years later, Boyden was first author on the first optogenetics paper, with Deisseroth as the principal investigator and last author. Wissner-Gross described the protein in that paper as bacterial rhodopsin. Karolinska's account of the prize, by contrast, centers on channelrhodopsin from an alga, the protein Hegemann and Nagel identified.
Wissner-Gross called his own role "maybe a little bit of steering or encouragement." He noted that Boyden did not share the prize, which went to Deisseroth, his principal investigator. Whether that is fair, he said, is a "separate story." His broader point was about the timeline: that "within a quarter of a century, it was possible to start from the sci-fi to the Nobel Prize for the sci-fi, I think is a remarkable case study."
Salim Ismail, founder of Open ExO, said both Boyden and Deisseroth had spoken at Singularity University. He welcomed the technology as a way of manipulating neurons. To people alarmed by the idea of "playing with the brain," he answers: "we have an old word for this. We call it marketing." Advertisers already try to provoke responses in the brain, he said; when someone sees a Coke, "you want them to get thirsty." Switching neural circuits on and off, Ismail said, gives "this magical read-write capability, which we've always wanted for the brain."
Should the Nobels change?
Asked for his view, Emad Mostaque, founder of Intelligent Internet and the episode's guest, agreed that Nobel Prizes take years and years. He went further: apart from economics, which he set aside, the prizes now "have to be for applied science." On purely theoretical work, he argued, the committee cannot keep up, or tell how much of a result came from AI and how much from humans; he mentioned AlphaFold among his examples. He proposed keeping the Nobel for applied science and adding another prize for what humans and AI achieve together. Such awards, he said, should go "to everyone, not just the PI," the principal investigator who leads a lab.
Dave Blundin, founder and general partner of Link Ventures, suggested the Nobel "might just become super quaint." He compared it to a historical village where visitors watch people churn butter. In 20 years, he imagined, the committee will be saying, in effect, remember when people had to think through this stuff manually, and "five people will show up." Diamandis said it was hard to believe that future Nobel Prizes would not all be AI-derived.
Wissner-Gross said that shift has already begun, citing Demis Hassabis's prize for AlphaFold and Geoffrey Hinton's prize for work on Boltzmann machines, an early type of neural network. He questioned the second award, saying he did not know anyone who uses them. Blundin said Hinton "absolutely deserved the Nobel Prize for backprop," meaning backpropagation, the method used to train neural networks, but that work is not physics and so was not eligible. In his view, the committee looked for something else Hinton had done and settled on the Boltzmann machine. Blundin called the choice "contrived," and Wissner-Gross agreed it was "a weird one." Blundin's explanation: "They wanted to give a Nobel Prize for AI." Wissner-Gross said he expects many more AI prizes in physics and chemistry.
A brief turn to physics
The physics prize was announced on the morning of the recording. It went to Francis Halzen of the University of Wisconsin–Madison for his role in the IceCube Neutrino Observatory and the discovery of high-energy neutrinos from space. The university's announcement describes IceCube as an international collaboration that has embedded thousands of light sensors in a cubic kilometer of Antarctic ice. The sensors catch rare flashes from neutrino interactions. Because neutrinos travel largely undisturbed, they can help identify distant sources when combined with other astronomical observations.
Wissner-Gross saw another limit of the Nobel format here. Prizes have to go to a person, he said, but projects like IceCube or CERN are "massive organizations with lots of people," so the award typically goes to the lead. He also made a larger claim, acknowledging he might "get in trouble for this": fundamental physics has had a noticeable deficit of advances in the past 50 or so years. With a deficit like that, he argued, the committee is "stuck awarding it to applications," a point he tied to Mostaque's argument. He then said he would "talk my book for a few seconds": he co-founded Physical Superintelligence partly to revive fundamental physics and bring about "a second physics golden age."
Diamandis closed the topic with a fun fact: every second, he said, 65 billion neutrinos from the sun pass through you without you noticing. Ismail replied: "I did not know that. And I'm not sure I needed to know that."