In the spring of 1953, a 25-year-old named James Watson stood up in a lecture hall on a hill above a harbour on Long Island and told a room full of the world's best biologists what life actually looks like. A spiral staircase. Two strands of code winding around each other. He had the shape of DNA, and standing in that room, he quietly rewrote the story of what we are.

The shape was not Watson's alone. The clearest look anyone had at that spiral came from Rosalind Franklin, a crystallographer at King's College London, whose X-ray photograph of DNA was the image that made the structure legible. She died of cancer at 37, four years before a Nobel that, by its own rules, never goes to the dead. So before we go any further, a tip of the hat to her, and to every woman whose fingerprints are on this story without her name in the citation.

Seventy-three years later, in that same corner of the world, they did it again. Except this time the frontier wasn't a molecule. It was a machine that could read the molecule.

Let me back up.

There's a place called Cold Spring Harbor Laboratory. Wind through a quiet wooded suburb outside New York City, reach the water, and on the hill above it sits a cluster of white colonial buildings I can only describe as a temple of science. That's not hyperbole. Eight Nobel Prizes. Normalise for size and the Nature Index once ranked it the most prolific biomedical research institution on the planet. Watson ran the place for decades. Biologists call it the crossroads of biology, a polite way of saying that if you want to know where the field is heading, you go and stand at the intersection and watch what comes through.

I've stood there. During my PhD years, back when I was teaching machines to read cancer out of DNA, I made it to ‘The Lab’, and the stories don't oversell it. A gazebo on the walking trail with a virus carved on top. A sculpture called the Waltz of the Polypeptides. Most people never leave campus for the entire week. You eat there, you sleep there, you argue about chromatin over dinner and drink from what everyone calls the scientific firehose until your brain aches. It is my kind of monastery.

I even got to meet Watson himself on one of those visits. He was every bit as sharp as his legend, and by the end of his life he had been disgraced, stripped of his titles by this very lab for what he said about race. Both of those things are true, and I've never quite known where to file the memory.

Since 1933, Cold Spring Harbor has run an annual symposium, and every year they point it at whatever they think is the true frontier. In 1953 that frontier was the double helix. This spring, for the 90th symposium, they pointed it at something that has been quietly eating the whole of biology from the inside.

Artificial intelligence.

Now, before your eyes glaze over (I know we've all reached peak AI-in-a-headline fatigue), sit with me for one number. There's a line from Sydney Brenner, one of the founding giants of molecular biology, that has haunted the field for years. He said we are "drowning in a sea of data and starving for knowledge." For half a century biology has been extraordinarily good at generating data and comparatively hopeless at understanding it. We could read the letters. We just couldn't read the story.

That equation is what's changing. And the best way I can show you is through three scientists who each got up on that stage and, in their own way, made the room gasp.

The first was Jennifer Doudna, who won a Nobel for CRISPR, the gene-editing tool that lets us rewrite the code of life the way you'd fix a typo. She opened the whole thing. And she said something that reorganised how I think about her own invention. She called CRISPR a bioinformatic discovery. Not a biological one. Nobody found CRISPR by peering down a microscope at a bacterium. Back in the early nineties, a Spanish PhD student named Francisco Mojica was reading the genome of a salt-loving microbe he'd fished out of the marshes near Alicante when he noticed the same chunk of DNA stuttering over and over, like a single word repeated down a page. He had no idea he was staring at what would become one of the most powerful tools in the history of medicine. It took years of squinting at databases before anyone worked out what those repeats were for. In other words, computer science had already revolutionised biology once, quietly, decades ago, and most of us didn't clock it at the time.

Then she came full circle. Her lab went hunting for CRISPR's ancestors, the systems that came before it in the deep history of life, and this time the search was powered by AI. They sifted through hundreds of millions of protein structures and pulled out something older than CRISPR, a genetic system nobody had described, with its own hidden recognition code. They couldn't crack the code by hand. So they used a language model, the same basic technology that finishes your sentences in an email, except trained on the language of genes, and it read the pattern for them. Her line for where this all leaves us was pragmatic and perfect. "Big biological questions will need AI-assisted scientists."

The second scientist I want you to meet is Edward Buckler, and he got up and did the bravest thing you can do at a conference full of people obsessed with the human body. He talked about corn. "So why should we care about plants?" he joked. "I'm the only plant guy talking here." The room applauded, because the vibes all week were apparently excellent, and because he was right.

Consider the plants for a moment. They are 80% of all the biomass on Earth. Every animal, every fungus, every one of us, is a rounding error stacked on top of them. And over the last century, using genetics and better farming, American corn and dairy yields have gone up sevenfold. Seven times more food from the same ground, one of the most underappreciated wins in the history of the human race, and almost nobody talks about it. Buckler's new problem is harder and more urgent. He wants to predict how roughly 2,000 critical plant species will cope with a climate that no longer resembles the one they evolved in. His team built AI models that can read a plant's genome and forecast how it might adapt, ran them across more than 700 species of grass, and surfaced 17 genes that look like they hold the keys to survival. That's not a lab curiosity. That's a map for feeding people through a century of heat.

And the third. If you only remember one thing from this newsletter, make it this one.

There's a researcher in Vienna named Andrea Pauli who studies fertilisation. The single moment when a sperm cell and an egg cell recognise each other and fuse, the event that every animal that has ever lived is the downstream consequence of. She calls it life's first kiss. And this is the quietly astonishing part. We didn't really know how it worked.

The scientists who study this clearly feel the romance of it. The key protein on the sperm is called Izumo, after a Japanese shrine where couples go to be married. The egg protein it reaches for, identified years later, was named Juno, after the Roman goddess of marriage. Two molecules named for wedding vows, groping toward each other in the dark. Naming them turned out to be the easy part. How they actually locked together was a mystery that had resisted the microscope for decades.

So Pauli's lab stopped trying to see it and started trying to predict it. They turned to AlphaFold, the AI system that can guess the three-dimensional shape of a protein from nothing but its raw sequence of letters. It has now done this for more than 200 million proteins, very nearly every one known to science, which is the kind of thing that wins you a Nobel Prize, as it did in 2024. Pauli's team ran the fertilisation proteins against each other inside a computer and asked a simple question. Which ones fit? The model handed back a structure. A specific molecular handshake, the bridge that joins sperm to egg, assembled from pieces nobody had been able to fit together at the lab bench. In his Nobel lecture, Demis Hassabis singled this out as one of the most remarkable things AlphaFold has ever done. Pauli's team has since gone further and modelled the entire assembly of proteins that makes the whole thing possible.

Now, the careful scientist in me has to tell you something, because Pauli was scrupulous about it herself. This is a prediction, not a photograph. Nobody has purified all these proteins and physically watched them click. But her lab ran experiment after experiment showing that every single one of those proteins is necessary, and that they all seem to work as a unit. The description that stopped me cold was this. The AI is behaving like a digital microscope that sees further than any real microscope we own.

Read that again. We built a lens out of mathematics that resolves things our physical instruments cannot. If you have ever heard me bang on about how the whole trick of being human is your ability to shift your perspective along a spectrum, from quarks to galaxies and back, this is that idea made real in a lab in Vienna. A new kind of lens. A new place to stand.

I should be honest about the mood in that room, because it wasn't all wonder. A lot of the week was about AI agents, systems that don't just answer questions but go off and do the science themselves, forming hypotheses, designing experiments, running for days. Google's team showed off an "AI co-scientist," and told a story about a microbiologist named José Penadés who had spent ten years working out how certain superbugs pull off a piece of genetic burglary, stealing tails from other viruses so they can break into bacteria they had no business infecting. He pointed the AI at the puzzle. It reached his answer in two days. His first reaction, understandably, was to fire off an email to Google asking whether they'd somehow been reading his private files. They hadn't. Worse, or better, it also handed him four more hypotheses that made sense, one of them something his lab had never even thought of. They're working on it now.

When the applause settled, a graduate student stood up and asked the only question that really mattered to him. What does this mean for me? For the students I teach?

Nobody in that room pretended to have a clean answer, and I'm not going to pretend for them. The reception all week was a very human cocktail of excitement, dismissiveness and quiet anxiety, sometimes in the same person, sometimes in the same sentence. Scientists have been promised self-driving laboratories for decades and most of the work is still done by hand at midnight by a tired postdoc. Extraordinary claims still need extraordinary evidence. But something has clearly shifted.

The best measure of that shift came from Ewan Birney, a legendary bioinformatician who helped map the human genome. For years he thought these AI models, with more knobs to tune than there were data points to tune them on, were basically voodoo. His words. This year he described himself as Saul on the road to Damascus. A convert. A believer, blinking in the new light.

Science does this every time. Barbara McClintock worked out that genes could jump around inside the genome while she was standing on that very campus, and the field ignored her for so long she quietly stopped publishing her data. Thirty years later they gave her a Nobel Prize, alone, at the age of 81. The same institution that once shrugged at her is the one that convened this. The heretics become the canon. They always do.

Here's what I keep coming back to. For most of its history, biology was a science of collecting. We collected specimens, then genes, then whole genomes, mountains and oceans of the stuff, all those letters, and we drowned in them exactly as Brenner warned. What happened at Cold Spring Harbor this year is that we finally started reading. Not the letters. The meaning. CRISPR's lost ancestors, the survival genes of grass, the molecular shape of the first kiss. Things that were always written into the data, waiting, that no human eye could resolve on its own.

Watson stood in that room in 1953 and showed us the alphabet of life. It took us the better part of a century to learn to read it. And now, in the same room, we're being handed a lens that can finally read it back to us.

The alphabet took a century.

Let's see what we do with the story.

If this is the kind of future you're trying to make sense of, it's the work I do. My book Full Stack Human is about staying human in a world run by technology, and biotech and the next era of medicine sit right at the centre of my Next Economy talk. If that's what your team is wrestling with, come find me.

Stay human. The world needs it.

— Tāne Hunter
Founder, Future Crunch
Co-author, Full Stack Human

P.S. A small piece of trivia I can't resist, because the universe has a sense of humour. Claude Shannon, the man who invented information theory and basically taught the world to think about data as data, wrote his PhD thesis on the mathematics of genetics. And he wrote a good chunk of it at Cold Spring Harbor. The idea that life is information, and information is something we can finally learn to read, was quietly germinating on that hill long before any of us showed up with our machines. The thread runs deep.