“A good A.I. future cannot be one where a small number of institutions control most of the capability and most of the upside. It should be a future where many people, companies, communities, and countries can build, benefit, and hold power. We believe this transformation should belong to everyone.”
—OpenAI, sharing its Lehane-ian plan to make sure A.I. benefits everyone. The document is full of the kind of rhetorical 180s that OpenAI has been making for the past few weeks—an interesting contrast with Anthropic’s more restrictive Fable 5 launch.

Runner-up: “Our highest and most urgent national priority should be A.I. safeguards. The risks of A.I. weapons, pathogens, mass unemployment, surveillance, and even extinction must not continue to be largely ignored.”
—Mitt Romney, who sounds like he read the pope’s first encyclical, weighing in.

And now for the main event…

Jeff Carroll was in the Army when his mom was diagnosed with Huntington’s disease—a fatal, hereditary, neurodegenerative disorder that Carroll, a high-school dropout, had never heard of. So when he got out of the Army, he enrolled in a few university biology classes to better understand it. Huntington’s, he told me—unlike other neurodegenerative diseases such as Alzheimer’s or Parkinson’s—results from one specific genetic mutation, which the research community has known about since 1993. Carroll himself carries the mutation. “You would think, given all that knowledge, or all that time, we’d have figured it out by now,” he said. “We know a lot, but not enough, because we can’t cure anyone of H.D. yet.”

Carroll, in many ways, has devoted his life to the disease. He’s spent the last 15 years researching Huntington’s, and picked up his Ph.D. in neuroscience along the way. More recently, he’s become one of the lead investigators for Huntington’s research at the Brain Health accelerator, a massive global effort announced last week by the Allen Institute, the nonprofit medical research organization founded by the late Microsoft co-founder and polymath Paul Allen.

Brain Health, which is focused on treating neurodegenerative diseases, is launching with a $400 million funding commitment—half of which is coming directly from the Allen Institute, in addition to $100 million from the Bezos family and $100 million from A.W.S., the National Institutes of Health, and EverythingALS. The initiative will focus first on Alzheimer’s, Parkinson’s, Huntington’s, Lewy body dementia, and amyotrophic lateral sclerosis (A.L.S.).

Neurological diseases have been particularly vexing for medical researchers for a myriad of reasons: namely, the complexity and inaccessibility of the human brain, the blood-brain barrier that blocks many therapeutic drugs, and technical and ethical limitations to conducting experiments on humans. Only recently has technology—fueled by advances in A.I.—begun to catch up to researchers’ ambitions. The Allen Institute already had one of the largest collections of high-resolution brain imaging data in the world—a critical first step, researchers hope, to putting new A.I. models to work. “The tools have gotten far enough now,” Carroll said, “that they’re productive across the scale of hundreds of human brains, which is what you need to see these early changes.”

Since the term “artificial intelligence” truly entered the zeitgeist in 2023, the leading labs have vowed that any of the technology’s risks would be offset by world-changing benefits—perhaps none more so than the potential to cure disease. “Many of my friends and colleagues are raising children, and when those children grow up, I hope that any mention of disease will sound to them the way scurvy, smallpox, or bubonic plague sounds to us,” Dario Amodei wrote in his 2024 “Machines of Loving Grace” essay.

He’s not the only one to have evoked this quixotic future. A little more than a year ago, Google DeepMind chief Demis Hassabis told 60 Minutes that our society would soon be able to “cure all disease with the help of A.I.” Sam Altman was saying much the same as recently as March, promising that “A.I. will help discover new science, such as cures for diseases, which is perhaps the most important way to increase quality of life long-term.”

The folks at the Allen Institute are optimistic that the Brain Health project will yield treatments for neurological disorders, but they’re also intensely aware of the work ahead. “I don’t think it’s going to be a magic box,” Carroll said, adding that it “remains to be seen how much these techniques can push beyond what we already know.” Mariano Gabitto, who leads algorithmic development for the project, said, “It’s a huge endeavor. We don’t plan to solve it at once.”

Still, the undertaking sounds like something out of a movie. The project is split into several main pillars, one of which Gabitto described as an “open science discovery engine” whose goal is simply to better understand the brain—an enormous task that involves the complex process of brain data collection, curation, and study. “You actually have to understand the components. You have to understand what goes wrong with them,” Ed Lein, the project’s director, told me.

Treatment is another pillar of the project, which also aims to produce therapies that target individual cell types, using different forms of A.I. and machine learning technologies to analyze data, identify patterns, and highlight potential candidates or patterns for early detection. “A.I. is at the center of everything,” Gabitto said, describing the creation of models designed to predict a disease’s trajectory, which could enable doctors to identify not just its type, but its severity and progression.

Researchers and donors hope this confluence of approaches will take them much further than other moon-shot efforts to better understand the brain. Gabitto called it “precision biology,” which he said would ideally translate into “precision medicine.” Part of the breakthrough, according to Lein, involves getting foundation A.I. models “to learn the structure and the grammar, if you will, of biology”—a potentially game-changing leap beyond the machine learning that has already become so thoroughly integrated into neuroscientific research efforts.

Lein expects the first few years of the project to be “catalytic,” as various communities rally around the initiative. “If we can solve the technology side, it’s going to open up the biology side—and, for that matter, the A.I. side,” he said. “So I think we’re going to see a much larger community that wants to join in.” But all three researchers expect their progress to take time, which is not something many people with these diseases have. “In some way, I try to disconnect my survival from the everyday lab work, because otherwise it gets kind of overwhelming,” Carroll told me. “You have to just trust that this worldwide team of folks are pulling in the right direction. And that’s what gives me hope.”

That’s all for today. I’ll see you on Tuesday.

Ian

Need help? Review our FAQ page or contact us for assistance. For brand partnerships, email ads@puck.news.

You received this email because you signed up to receive emails from Puck, or as part of your Puck account associated with {{customer.email}}. To stop receiving this newsletter and/or manage all your email preferences, click here.

Puck is published by Heat Media LLC. 107 Greenwich St., New York, NY 10006