Skip to content
a16za16z

Taking Bold Bets: NIH and the Future of Biomedical Science

Dr. Jay Bhattacharya is one of the country’s top medical experts and a 24-year professor of medicine at Stanford. After being censored and deplatformed during COVID for his role in opposing harsh lockdowns, he was appointed Director of the National Institutes of Health by President Trump in 2025. a16z General Partners Erik Torenberg, Vineeta Agarwala, and Jorge Conde join Dr. Bhattacharya to discuss the administration’s role in tackling the autism crisis, how to restore public trust in health authorities, how to make the NIH more dynamic and efficient, and how to streamline publishing and restore academic freedom. Timecodes: 00:00 Introduction 00:59 Autism Research Initiatives Announced 02:01 New Findings: Leucovorin and Tylenol in Pregnancy 04:40 Addressing Preterm Birth & Broader Health Concerns 06:10 The Replication Crisis in Science 09:26 NIH Funding, Grant Review, and the Silicon Valley Spirit 12:47 Grant and Review Process 14:26 Portfolio Management & Allocation at NIH 26:23 The Challenge of Supporting Early Career Investigators 31:23 Allocating Grants and Training Under the New Administration 35:46 Academic Freedom and Scientific Publishing 38:34 Rebuilding Public Trust in Science & Public Health 41:57 Communicating Uncertainty & Scientific Honesty 48:58 NIH Priorities: Nutrition, Chronic Disease, and AI 54:15 Advice for the Next Generation of Scientists 56:37 The Role and Limits of AI in Science Resources: Find Dr. Bhattacharya on X: https://x.com/DrJBhattacharya and https://x.com/NIHDirector_Jay Find Erik on X: https://x.com/eriktorenberg Find Jorge on X: https://x.com/JorgeCondeBio Find Vineeta on X: https://x.com/vintweeta Learn more about the NIH: https://www.nih.gov/ Stay Updated: If you enjoyed this episode, be sure to like, subscribe, and share with your friends! Find a16z on X: https://x.com/a16z Find a16z on LinkedIn: https://www.linkedin.com/company/a16z Listen to the a16z Podcast on Spotify: https://open.spotify.com/show/5bC65RDvs3oxnLyqqvkUYX Listen to the a16z Podcast on Apple Podcasts: https://podcasts.apple.com/us/podcast/a16z-podcast/id842818711 Follow our host: https://x.com/eriktorenberg Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures.

Dr. Jay BhattacharyaguestErik TorenberghostVineeta AgarwalahostJorge Condehost
Sep 22, 202558mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

NIH’s bold-bet reset: replication, risk-taking, trust, and AI integration

  1. NIH is launching a $50M Autism Data Science Initiative (13 teams selected from ~250 applicants) aimed at generating clearer answers for families amid rising autism prevalence.
  2. Bhattacharya highlights emerging clinical signals—leucovorin (folinic acid) helping a subset of autistic children and new evidence linking acetaminophen use in pregnancy to later autism risk—framing both as areas for cautious guidance and further study.
  3. He argues the “replication crisis” stems from science’s difficulty, specialization, weak incentives to replicate, and peer review/publication standards that mistake publication for truth.
  4. He advocates importing a Silicon Valley-style portfolio mindset into NIH funding: tolerate productive failure, fund newer ideas, and judge success at the portfolio level rather than per-grant certainty.
  5. To rebuild trust after pandemic-era overconfidence and inconsistent policies, he calls for “gold-standard science,” humility about uncertainty, academic freedom, and open access publishing—while positioning AI as an augmenting tool that still requires careful validation.

IDEAS WORTH REMEMBERING

5 ideas

Autism work is being reframed as a data-science and systems effort.

Bhattacharya describes autism prevalence as high and rising and says NIH lacks actionable causal answers; the new $50M initiative is designed to fund teams capable of integrating datasets and producing clearer, testable hypotheses over the next few years.

Leucovorin is presented as a targeted opportunity, not a universal autism “fix.”

He claims folinic acid can help children with specific folate-processing issues, citing effects like speech restoration in a minority and broader improvement in others, while emphasizing that benefits depend on subtype/biology and should be made more accessible via coverage and guidance.

Pregnancy acetaminophen guidance is moving toward prudence amid contested evidence.

He references newer studies suggesting correlation with later autism diagnoses and supports updated FDA guidance that encourages limited, need-based use (e.g., high fevers) without “panic,” reflecting uncertainty rather than definitive causality.

Replication is treated as the core “truth test” that current incentives underfund.

He argues scientists gain little career reward for replication and negative results often go unpublished; raising standards and funding independent verification would reduce false confidence and improve downstream translation to health benefits.

NIH wants to act more like a portfolio investor—accepting failure to buy breakthroughs.

Bhattacharya contrasts venture portfolios (many failures, a few transformative wins) with NIH’s increasing conservatism, noting funded ideas have aged over decades; he proposes tolerating “productive failure” and enabling publication of what didn’t work and why.

WORDS WORTH SAVING

5 quotes

The standard for truth in science ought to be replication. Independent teams. You don't-- Just don't believe me just 'cause I have an-- I say something is true. You know, other people independently looking at the same thing should arrive at the same answer.

Dr. Jay Bhattacharya

We should stop punishing scientists who fail. If they fail productively, let them publish in a journal to explain why they're-- what they, what they learned from it. Like, that Silicon Valley spirit, I think, needs to come to science a little bit more.

Dr. Jay Bhattacharya

The fact that it's published in a journal doesn't mean it's right. It doesn't mean it's true.

Dr. Jay Bhattacharya

The American people are not stupid. In fact, they're quite smart, and when we talk to them in ways where we show respect for, for their, their, their intelligence with d- with data, allow people to disagree, but then have the evidence right there in front of people, I think people, people will respond with trust, uh, w-where the evidence actually leads.

Dr. Jay Bhattacharya

It's the individual scientists who believes in their idea, you keeps knocking on the door even when the, the door is closed over and over again until it opens. That's, that's who really makes a big difference in this world.

Dr. Jay Bhattacharya

Autism Data Science Initiative ($50M)Leucovorin/folinic acid and autism subtypesAcetaminophen (Tylenol) guidance in pregnancyPreterm birth outcomes vs. EuropeReplication crisis and reproducibility incentivesNIH grant review centralization and portfolio managementEarly-career investigator pipeline and age of first major grantAcademic freedom and open-access publishingPublic trust after COVID-era policiesAI in research, clinical workflows, and grant process noise

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome