Overview
Decentralized Science (DeSci) is a movement to reform how science is funded, published, and governed using public blockchains, DAOs, and token incentives. As the Ethereum Foundation's DeSci hub frames it, the goal is to build public infrastructure for “funding, creating, reviewing, crediting, storing, and disseminating scientific knowledge” – addressing systemic failures of the traditional research system. EncrypGen, a genomic-data marketplace, claims to have coined the term: its co-founder David Koepsell wrote in May 2022 that the company believes its first public use of “DeSci”, modelled on “DeFi”, came in a YouTube AMA on 26 February 2021.
The problems DeSci targets
- Funding bottleneck – grant capital is scarce, slow, and gate-kept; low single-digit acceptance rates push researchers to optimize for grant committees rather than discovery.
- Closed-access publishing – publicly-funded results sit behind paywalls, and peer review is opaque and unpaid.
- IP lock-up – promising early research stalls in university tech-transfer offices, unable to attract capital or reach the public.
- Misaligned incentives – the reproducibility crisis and negative results are undervalued because they don't advance careers.
Core primitives
- IP-NFTs & IP Tokens – tokenize the legal rights to research so a community can own and fund it.
- DeSci DAOs – token holders collectively decide which research to fund and how to commercialize the results.
- Novel funding – quadratic funding, retroactive public-goods funding, and token-curated grants replace the single grant committee.
- Decentralized publishing & peer review – open-access research objects and on-chain attestations that give reviewers reputation and, increasingly, payment.
- Research data ownership – data NFTs, data unions, and Data DAOs that let the people who generate genomic and clinical data keep ownership of it and a share of the value.
Who runs the experiments a DeSci DAO funds?
Usually an existing lab, paid from the DAO's treasury. A DeSci DAO holds money and a vote rather than benches or instruments, so it buys the work. The first IP-NFT set the pattern in August 2021: VitaDAO funded a longevity project at Morten Scheibye-Knudsen's laboratory at the University of Copenhagen, and in return owns the IP and the rights to future data.
The other route is the virtual biotech model. Writing for a16z's Future in April 2022, Jocelynn Pearl described companies that give up their own lab space and buy the work as a service: from contract research organisations (CROs) such as Charles River Labs, which run preclinical and clinical studies to order; through sponsored research agreements, which pay an academic lab; or from cloud labs such as Emerald Cloud Lab and Strateos, automated labs a scientist runs remotely. Both cloud labs still sell that service: Emerald Cloud Lab describes software-controlled laboratories in which scientists “design, execute, and analyze experiments remotely”, and Strateos sells access to its robotic lab for drug discovery.
The same guide described LabDAO building a marketplace where small companies and academic researchers could hire microCROs for work such as bioinformatics analysis and automated cloning. LabDAO has since wound down. Reputable, whose documentation still describes a community-governed REPT token, now designs and runs studies for health and wellness brands with wearables, surveys, biomarkers and lab tests. It calls them “fully decentralized” and recruits from a community of more than 10,000 participants.
Why now: the 2025–26 public-funding shock
DeSci's funding-bottleneck thesis stopped being abstract in 2025. Over the year the U.S. government terminated or froze more than 3,800 NIH and NSF research grants – roughly $3 billion in unspent, already-committed funds, the NIH alone accounting for nearly 2,500 grants (about $2.3 billion). One analysis found the terminations disrupted roughly 1 in 30 active clinical trials, affecting more than 74,000 enrolled participants. When the dominant grant pipeline contracts this sharply, community- and token-based funding stops being a novelty and becomes a lifeline – the clearest demand-side tailwind DeSci has had.
The 2026 landscape
DeSci has consolidated around a few coordination layers. Bio Protocol coordinates ten BioDAOs, a count read off its own ecosystem listing on 15 October 2026 and unchanged since 11 September, through the rebrand described below. Its documentation still calls it “a new financial layer for funding and developing early-stage biotech”, wording its homepage dropped in that rebrand. It raised over $33M in its BIO Genesis round in November 2024, Binance Labs' first DeSci investment. Both figures are dated on a published timeline Bio has since removed from its site, and are read here off the Internet Archive’s 21 August 2026 capture.
The ten span longevity (VitaDAO), women's health (AthenaBIO), psychedelics (PsyDAO), synthetic biology (ValleyDAO), hair loss (HairDAO), cryopreservation (CryoDAO), brain health (Cerebrum DAO), rare disease (Curetopia), Long COVID (Long COVID Labs) and quantum bio-imaging (Quantum Biology DAO) – all ten covered here. The listing is not the whole cohort. DermaLabs came off the same launchpad: its $SKIN Ignition Sale ran in December 2025, and Bio published its own profile of the project in February 2026. It appears in neither the live listing nor the Internet Archive’s 12 June 2026 capture of it, so a count read off the listing and a count read off the launchpad’s own history differ by one.
At DeSci.Berlin in June 2026 it launched OpenLabs, backed by a fresh $6.9 million round. Read on 9 September 2026, openlabs.bio.xyz is not the funding surface press coverage implied: it bills itself as “Where Scientists & AI Agents Discover Breakthroughs” and invites researchers to “post hypotheses, challenge ideas through peer review, and collaborate with researchers and AI agents”. Funding still happens somewhere else – on app.bio.xyz/launchpad, which on the same date carries links to exactly the ten BioDAOs above and no eleventh.
Bio has meanwhile rebuilt its homepage around three further surfaces – Buildspace for getting a project funding-ready, Biofy for shipping products, and BIOS, an AI research assistant – retitling it “Science for the Network Age” and retiring both the “Gateway Drug to Decentralized Science” hero and the word “bioDAO”, which now appears nowhere on the page. The Internet Archive’s last capture of the old homepage is dated 21 August 2026 and its 6 October 2026 capture holds the new one, so the redesign landed in the three weeks after 21 August; none of Bio’s blog posts announces it. The direction is the one to note: DeSci's frontier is moving from tokenizing IP toward operating the research workflow itself, and it is doing so by splitting the workflow across surfaces rather than fusing it into one.
The homepage also links three case studies, read on 27 September 2026, and between them they show how Bio's launch terms have changed. The HairDAO case study, dated to HairDAO's launch in the first cohort in November 2021, describes an open auction in which buyers set the clearing price and no valuation was fixed in advance. It sets that against today's Ignition Sales, which open at a fixed fully diluted valuation, “currently in a $2M to $3M band”, against a raise size agreed before the sale.
The ARTAN Bio case study shows the IP-NFT route: $91,000 on a community vote in July 2023, the IP-NFT tokenized into five million tokens, a fifth of them sold for $300,000 in June 2024, and the IP itself kept by the company, so holders get a vote on the research rather than ownership of it. The PeptAI case study covers the first launch on Bio's coin-to-company framework, a $50,000 raise in May 2026 whose governance token has a stated route to convert into equity; Bio, which built PeptAI in-house, reports $112,000 of trading fees paid to the project since. That is three ownership structures from one launchpad: a DAO that owns its patents, a company that owns its IP, and a token meant to become equity.
Relevance to Radix
Radix's asset-oriented programming is well-suited to DeSci: IP-NFTs are native resources (not contract state), DAOs benefit from engine-level access rules, and complex funding flows compose atomically via transaction manifests. See DeSci on Radix.
How Caper approaches this
A research collective doesn't need a bespoke platform to run the DeSci playbook. A caper gives any group a bonding-curve treasury to raise from supporters and a proposal system to direct that treasury toward a lab, a milestone, or an IP acquisition – with a participation-weighted exit open to any contributor holding the caper's soulbound vote tokens – minted only by ranked ballots, one per ballot, so a backer who has bought but never voted has no exit to take – letting contributors who lose conviction withdraw their share rather than being locked in.
That share is not a flat slice by balance: exit prices the treasury payout with the same (t·v)/(V·T) canonical weight that decides a vote, so of two backers holding the same tokens, the one who has voted on more proposals redeems more of the treasury, and one who only bought redeems nothing through exit at all. It's the same “community funds the science, community governs the outcome” loop that DeSci pioneered, without a research group having to deploy its own contracts. See DeSci funding.