Decentralized science five years on: what DAOs fund, and what they have not yet shown

A researcher applying for a grant at the US National Institutes of Health has long faced odds of roughly one in five, and the wait between submission and money is often close to a year. In fiscal year 2025 those odds got worse. According to NIH data reported by STAT, the success rate for early-stage investigators applying for R01-equivalent grants, the agency's standard research award, fell from 29.8 percent in 2023 to 18.5 percent. The agency's budget itself held: Congress set it at about $48.7 billion for fiscal 2026.

That context explains some of the renewed interest in an idea that predates it. In 2021, a loose coalition of people from crypto and biotech began testing whether blockchain tools could fund research differently. They called it DeSci, for decentralized science. An episode of Tech Anatomy's ERROR 404 series follows the movement from those beginnings, describing it as either an important innovation or "a well-intentioned experiment still searching for proof that it can work at scale".

The problems DeSci set out to address

The criticisms of conventional research funding are older than DeSci and are shared by many people who have no interest in crypto. Grant writing takes months. Peer review panels are often described as favouring incremental projects likely to produce publishable results over riskier ideas. Researchers without an established record find it hardest to break in. Much of what gets published sits behind subscription paywalls, including work paid for by taxpayers.

Reproducibility adds a further question. In a 2016 Nature survey of more than 1,500 researchers, over 70 percent said they had tried and failed to reproduce another scientist's experiments.

None of this erases the record of the existing system. NIH peer review has funded much of the research behind treatments for diseases that were fatal a few generations ago, and other national agencies model themselves on it. DeSci's proposal was not a better review panel but a different mechanism.

How the mechanism works

The infrastructure came first. Molecule, founded around 2019 and 2020, developed the IP-NFT, short for intellectual property non-fungible token. The idea is to represent the legal rights to a research project, often before any patent is filed, as a token on a blockchain that can be owned, traded and governed by a group. In mid-2022 Molecule raised $12.7 million in seed funding, led by Northpond Ventures.

On top of that layer sit DAOs, or decentralized autonomous organisations. A DAO works somewhat like a cooperative in which members hold tokens rather than shares, and decisions are taken by on-chain votes rather than in a boardroom. VitaDAO, launched in 2021 around longevity research, was the first to draw wide attention. Its premise was that ageing research is underfunded, so a community could pool money, vote on projects and jointly hold the resulting intellectual property. By 2025 it had raised more than $10 million in total, according to an analysis in California Management Review.

In January 2023, VitaDAO closed a $4.1 million round in which Pfizer Ventures took part, alongside Shine Capital and others. According to VitaDAO, Pfizer was the first pharmaceutical company to vote on one of its proposals. The episode presents Pfizer's interest as a strategic bet on the "valley of death", the gap between academic research and commercial development, rather than an endorsement of crypto.

Other groups followed with their own focus. AthenaDAO funds women's health research and reports having deployed around $1.5 million across several labs. ValleyDAO works on synthetic biology. In November 2024, Binance Labs, the venture arm of the crypto exchange, made its first DeSci investment in BIO Protocol, which describes itself as an accelerator for these specialised "BioDAOs".

Two readings of the same record

The more favourable reading starts from a concrete case. In October 2023, VitaDAO took part in a $6 million funding round for Gero, a Singapore company using AI to look for drug targets linked to ageing. In July 2025, Gero signed a research and license agreement with Chugai, a member of the Roche Group, under which Chugai pays an upfront fee and up to about $250 million in development and sales milestones, plus royalties if a product launches. On this view, DeSci money can reach science that established industry later judges worth paying for, at a stage where conventional funders tend to hesitate.

The more cautious reading looks at scale and selection. The capital deployed across DeSci is measured in tens of millions of dollars, against tens of billions for the NIH alone. Projects cluster around longevity and biotech themes that appeal to the people most active in crypto communities, which raises the question of whether community funding follows funder enthusiasm more than unmet need. The episode asks where the DeSci effort for neglected tropical diseases or antibiotic resistance might be. On this view, one partnered company shows that the model can pick a promising team, not that it outperforms other ways of doing so.

Both readings fit the facts. They differ on whether DeSci should be judged as a replacement for public funding, which it is far from being, or as a niche source of early, risk-tolerant capital.

What the evidence does not settle

Several structural questions remain open. Governance usually rests on token-weighted voting, so influence over funding decisions can track token holdings rather than scientific expertise. VitaDAO and others add scientific advisers and review frameworks, and the episode calls the problem "not a fatal flaw, but it's a real flaw". Treasuries held in cryptocurrency shrank during the 2022 market downturn, and many DAOs have since moved reserves into stablecoins. The legal status of tokenised intellectual property is also unsettled. The EU's MiCA regulation on crypto-assets has applied in full since the end of 2024, but how an IP-NFT interacts with patent law, securities rules and research agreements is still being worked out.

A funding agreement is also not a clinical result. The Gero deal is contingent on milestones, and its full value is a ceiling, not a sum paid.

What would show whether the model works

The episode closes on a practical question: "Can DeSci produce a therapy that reaches patients?" Drug development takes a decade or more, and the movement is about five years old. The indicators worth following are whether DAO-funded projects enter and complete clinical trials, whether licensing deals such as Gero's reach their milestones, whether funding spreads beyond longevity into less fashionable diseases, and whether governance and reserve practices hold up through the next crypto downturn. Some of DeSci's ideas, faster decisions and transparent ownership of early IP among them, could also be adopted by conventional funders, which would make the movement's influence harder to measure than its own balance sheet.

Listen to the episode

ERROR 404 - The Story of DeSci

Listen · 17 min