My aunt kept a pill box on the kitchen counter. The morning I read OpenAI Foundation’s announcement, I was folding her laundry and felt the familiar, small panic that comes with losing someone to Alzheimer’s. You and I both know the difference money makes in a lab: it buys time, people, and the awkward, repetitive experiments that find answers.
At 9:12 a.m. a short blog post appeared on the OpenAI Foundation site — The OpenAI Foundation’s New Alzheimer’s Effort Helps Demystify ‘AI Will Cure All Diseases’; You know what speeds up medical research? Money.
The Foundation says it is “finalizing more than $100 million (€93 million) in grants this month, across six research institutions, to support and accelerate Alzheimer’s research.” That line reads like a headline and a promise at once. I want you to hold two things in mind: funding changes timelines, and funding pays people to do the slow work AI can’t do alone.
On the Foundation’s page I found five research goals listed and a note on brand language — the list reads like a research brief, not a manifesto.
The Foundation frames its work around five aims:
- Create a causal map of Alzheimer’s using AI to validate targets for intervention.
- Design new drugs with AI, and test them in the lab.
- Support open datasets to predict drug activity, and chart disease progression with and without intervention.
- Establish new biomarkers to improve diagnosis and how clinical trials are run.
- Test off-patent treatments and use AI to make sense of anonymized patient data and experiences reported online.
Four of those five explicitly invoke AI. Read generously, they’re sensible aims: better biomarkers speed trials, shared datasets cut duplicated work, and validated targets make drug discovery less random. But AI is not a substitute for clinicians, trial infrastructure, or patients who consent and show up for months of follow-up.
How much is OpenAI funding Alzheimer’s research?
Short answer: more than $100 million (€93 million) in initial grants, distributed across six institutions. That funding is earmarked for work that includes clinical trials — not just models in a server farm.
At an industry briefing last week the word “mission” came up again — and that’s where philanthropies differ in tone and transparency.
Compare OpenAI Foundation’s explicit grant amounts with Mark Zuckerberg and Priscilla Chan’s Biohub messaging. Biohub’s homepage declares “Our mission is to cure or prevent all disease.” When Chan and Zuckerberg’s foundation bought an AI biology lab, EvolutionaryScale, Axios reported the terms were not disclosed. The silence around dollars feels like a choice: opacity makes a project sound grander than it is.
OpenAI CEO Sam Altman has tweeted that “AI will help discover new science, such as cures for diseases.” That’s a credible point of view. It’s also easy to read as a broad claim if you remove the people who will do the work from the sentence.
AI will help discover new science, such as cures for diseases, which is perhaps the most important way to increase quality of life long-term.
AI will also present new threats to society that we have to address. No company can sufficiently mitigate these on their own; we will…
— Sam Altman (@sama) March 24, 2026
Can AI cure diseases?
Not by itself. AI can accelerate target identification, help design molecules, and improve trial matching — but the real work happens when those computational leads hit a lab bench and then human trials. Think of AI as a lens that points researchers at promising directions, not as the origin of the cure.
In a lab I visited, a scientist pipetted samples while a colleague scrolled model outputs on a laptop — the human-machine workflow was obvious.
Funding moves that workflow faster. Money is oxygen to labs: it pays salaries, reagents, longer trials, and the patient recruitment teams that clinical studies need. OpenAI Foundation’s grants promise to underwrite the human labor that turns model hypotheses into testable chemistry and measurable clinical outcomes.
That said, philanthropy framed as a brand can drift toward magical thinking. Meta-linked projects like Biohub, and public statements by founders, make it easy to imagine AI as a wand that will make diseases disappear overnight. AI can be a powerful tool, but it often amplifies existing problems — bias in datasets, limits in experimental design, and the unpredictable biology of the brain.
The OpenAI Foundation recently renamed its internal section from “Health & Curing Diseases” to “Life Sciences & Curing Diseases,” a small linguistic nudge that puts biology first. That matters: language shapes expectations for funders, scientists, and patients.
How will AI help Alzheimer’s research?
AI helps by testing hypotheses faster in silico, suggesting molecules for chemists to synthesize, improving patient stratification in trials, and spotting subtle patterns in imaging and biomarkers. Tools from companies and labs — including platforms like EvolutionaryScale and datasets referenced in journals such as ScienceDirect — often form the backbone of that work. But the gains show up only when models feed experiments and those experiments are conducted well.
At the end of the day you pay people and they get results — that truth undercuts a lot of the hype.
I have family who didn’t live to see better treatments. I’m not persuaded by slogans alone. If AI points to a molecule, it’s the chemists, clinicians, and trial participants who will prove whether it prevents a patient from losing another month of memory.
AI is a microscope that sometimes misses what hands reveal — and if you want rapid progress, you fund the microscopes, the labs, and the people who use them.
Gizmodo reached out to Biohub for comment, and we will update this piece if they respond.
The OpenAI Foundation’s move matters because it pairs language with an explicit dollar figure and a willingness to fund the messy parts of science. That recalibrates the story from “AI as miracle” to “AI plus people and capital.” Will that be enough to change how quickly Alzheimer’s yields to treatment, or will the headlines race ahead of the trials?