DeSci: The Intersection of AI, Crypto, and Life Sciences
TechFlow Selected TechFlow Selected
DeSci: The Intersection of AI, Crypto, and Life Sciences
Life sciences have moved beyond the meme fat tail.
Navigating Web3 tides with focused insightsAuthor: Zuo Ye
The black box of large models drives people crazy; blockchain hopes to bring transparent white-box practices to scientific research.
In 1943, Schrödinger—the man behind the quantum cat—delivered a high-level lecture in Dublin, arguing from statistical physics the relationships among atoms, life, and cells. At that time, a 15-year-old boy named Watson was already a freshman at the University of Chicago across the Atlantic.
After reading Schrödinger’s lecture-turned-book *What Is Life?*, Watson decided genetics would be his life's mission.
Ten years later, when Watson—who by then held a Ph.D.—proposed the double helix structure of DNA, the 25-year-old had already secured himself a Nobel Prize.
Grafting, Cloning, and Gene Editing
There are two trees in front of my house: one is a jujube tree, the other also a jujube tree.
Anyone who’s gone through middle school knows genes are information segments of DNA, like the "function body" in code—the most basic functional unit—while DNA resembles an instance module, and RNA acts as routing and communication, transmitting genetic information to specific targets.
Watson discovered DNA’s structure, but humans didn’t know how to use it—much like knowing about Schrödinger’s cat: cats are easy to find, but quantum communication still took decades to develop.
At least Watson was luckier than Schrödinger. In summer 2012, Emmanuelle Charpentier and Jennifer Doudna discovered that CRISPR sequences and Cas proteins could be combined: by artificially cutting specific sequences and inserting desired information fragments, the human body’s repair mechanisms could then seamlessly complete the gene editing process.
It really resembles gardening and pruning—visually connecting branches without understanding biological mechanisms, yet learning plant compatibility through repeated experiments alone.
Experiments continue, and so does cloning: nuclei and cytoplasm can be separated and “connected.” Through endless trials, cloning achieves isomeric wonders, just as described in *What Is Life?*
Gene editing isn’t mysterious—it’s simply cloning taken further, miniaturized to a microscopic scale. From an atomic perspective, life is ultimately nothing but the irreversible cooling down of thermal motion, much like time: perhaps stretchable or compressible, but never reversible.
Humans can graft fruit trees, clone animals—so can humans edit humans?
In 2018, the rogue scientist He Jiankui became either Eve or the serpent, editing the genes of twin embryos whose parents had HIV. Humanity thus opened Pandora’s box. Cloned animals can be humanely destroyed—but are gene-edited humans still human?
Image caption: CRISPR-Cas9 working principle, image source: @zuoyeweb3
But deeper exploration into genes holds fatal allure for certain groups—longevity. Find the gene segment affecting lifespan and, like General Jin the hacker, change its value from 100 to ♾️. Even adding a single zero might be enough.
In 2023, Paradigm co-founder Fred Ehrsam decided to leave the crypto industry and founded the biotech research company Nudge. Coincidentally, Fred was also a co-founder of Coinbase, transitioning into crypto VC after the company’s 2017 IPO.
Also in 2017, Paul Kohlhaas joined Consensys as BD lead, but left a year later to start his own venture. Why not use blockchain for something more meaningful?
For example, scientific research. In 2018, Molecule was founded—one of the earliest projects exploring the integration of blockchain and science, especially bioscience. Meanwhile, DeepMind—the parent company of AlphaGo—had already released its life sciences model AlphaFold in 2016, demonstrating strong capabilities in protein structure prediction.
In 2020, AlphaFold2 successfully solved the protein folding problem. Watson booked his Nobel at age 25; this time, the 4-year-old AlphaFold2 had already claimed 50% of the 2024 Nobel Prize in Chemistry.
Fred’s 2023 career shift wasn’t even early. As early as 2020, another Coinbase founder, Armstrong, launched ResearchHub, aiming to dismantle the institutionalized research pipeline of university-paper-funding, introducing incentive mechanisms to remove the three burdens—academic rank controlled by universities, profits captured by publishers, and funding applications—from scholars’ shoulders.
Especially problematic: researchers pay out of pocket to submit papers to publishers, while peer reviewers selected by publishers often work for free—only publishers profit in the middle.
All elements are now aligned: AI, research, and publications are converging on life sciences. Indeed, the 21st century is truly the century of biology.
Crypto Seeks Immortality, Starting with the Elixir of Life
Decentralized Science (DeSci) is pharmaceutical R&D under the banner of life sciences research.
DeSci is the crypto-world adaptation of the AI4Sci movement, highly focused on AI, life sciences, and new drug development. Perhaps it briefly wandered down meme-fueled detours—remember Paul Kohlhaas’s Molecule? It even received investment from Balaji in 2022. Who can resist the temptation of longevity?
Going further, in 2022 Paul Kohlhaas founded Bio Protocol, beginning development of products designed to extend the lifespans of crypto elites. It operates multiple sub-DAOs, covering scientific mysteries from head to toe, across all aspects of life.
In 2024, a "reborn" CZ and Vitalik appeared together at Bangkok’s DeSci Day, where young V—now V God—recommended to elder CZ the supplement VD001 from Vita DAO under Bio Protocol.
Then, Bio successfully secured investment from CZ’s YZi, its token smoothly listed on Binance, and Paul Kohlhaas proved adept at marketing, even launching a PumpFun-inspired Pump Science. Can memes combined with scientific research actually work?
However, after Bio’s surge came disappointing deliverables. In traditional research, developing a new drug routinely costs over $1 billion and takes years or even decades. But Bio’s secondary market can’t wait five minutes—taking funds without pumping the token makes actual research a crime.
The story isn’t over, because the Agent wave has arrived. AI Agents genuinely promise to boost research efficiency. More interestingly, ResearchHub received a $2 million Boost investment in February 2025, and now even Agent-generated DeSci papers undergo peer review.
In August 2025, Bio Protocol launched its V2 plan, unveiling a new Launchpad, the BioXP points program, and BioAgents built on ElizaOS—once again riding the latest trends.
Within just seven days, over 100 million BIO tokens were staked—80 million of them on August 7 alone—so the data remains somewhat questionable. That said, the V2 plan features a more sound economic design.
Small market cap avoids sell pressure and encourages continuous project sponsorship.
Image caption: $BIO staking data, image source: @cl2pp
Still, DeSci represented by Bio Protocol lags behind AI4Sci. AlphaFold open-sourced its database in 2021 and has since revealed 200 million protein structures, covering nearly all known species.
Feeling behind schedule, Bio Protocol has long hoped the U.S. FDA would release or consolidate big pharma’s accumulated data to accelerate open scientific research.
Besides, Bio V2 plans to launch multiple new drugs in the UAE, drastically shortening traditional R&D timelines. The Middle East’s lenient human trial regulations will further speed up life sciences research—though whether the outcome produces more He Jiankuis or more Watsons remains uncertain.
Conclusion
GPT-5’s performance was disappointing, but in specialized domains like medicine and scientific research, we can still await the flowering of Scaling Law. The data potential in these high-value fields hasn’t been fully tapped; once progress occurs, it could dramatically elevate human understanding.
In life sciences, Silicon Valley’s Colossal continues advancing its de-extinction project using CRISPR-Cas9 technology—such as the “hairy mouse” created by combining mammoth and mouse genes, or the massive pure-white terror wolf bred from ancient wolves.
Someday, humanity may evolve. Someday, humanity may go extinct.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
@zuoyeweb3
