Wisdom in fragments

Trust is a confident relationship with the unknown. And the reason this is so important is that trust isn't, at its essence, an asset or an attribute or a currency. It's a belief. It's what we believe about someone or something. That definition is slightly counterintuitive because, often, when people think about trust, they talk about knowing what to expect from someone or knowing what the outcomes will be. So, they're actually thinking about trust through the language of risk, thinking about it in terms of certainty. And that's not what trust is.

Author & Leading Expert on the Sharing Economy & Collaborative Consumption

We must always remember, in the West as democracies, that we have to do this within the character of a free society. That will hopefully always be what separates us from them: that we defend ourselves, but we do so within the character of a free society.

The idea that Google or Meta weren't American companies was a fiction they liked to tell themselves. It just simply wasn't true. And I'll tell you why: physically, their servers are in actual places.

Our brains are constantly making predictions about the world, and we learn when those predictions turn out to be wrong. So what Marshall did was deliberately redesign training so that people learned how to cope with a wide range of unpredictable challenges—turning the unpredictable into something more predictable.

Training helps you turn fear from something that paralyses you or makes you flee into something that spurs you to take the right decisions in very difficult situations.

You can have the best AI in the world and the best robots in the world, but if they aren't integrated well with the humans, then you will lose.

I think this is something that would have been a lot less controversial just a few decades ago, when people still remembered that, yes, the material side of war is obviously important, but it's only ever one aspect of it. And so, I'd say the big difference here is that we have to recognise that if we ignore or downplay the human side, we could lose.

There's a compelling mission—both because of the enormous economic opportunity it represents, if it succeeds, and because of the broader impact it can have on the world. The mission is defined by a stubborn technical challenge. It has to be hard. No one wants to work on something that isn't difficult, and in a sense, choosing the hardest problems becomes a moat.

Everyone on the team strongly believes in the mission. This is the one thing we want to work on in our lives, and multiple people share that same sentiment. We've fallen in love with a very stubborn technical challenge. Convincing others is always hard—it requires conviction on both sides. What we've demonstrated is small-team speed: speed in hiring, speed in getting strong results, and speed in refining our intellectual understanding.

At Axiom we've raised $64 million—we're a small startup—and we recently won the Putnam competition. We scored 90 out of 120, which would have placed us above all ~4,000 human contestants last year and at the level of a Putnam Fellow, meaning top five in the world. If you tried to achieve that purely through informal methods—where hallucination is a persistent risk—getting to the same level of consistent correctness would likely require a lot more resources.

A model that's really strong at mathematical reasoning is likely to be strong at coding. And a model that's excellent at both math and code is often very good at analysing the nuts and bolts of legal reasoning as well. The third and deepest reason this matters is the ability to bridge different levels of abstraction. All of these domains involve multiple layers of abstraction, and the ability to move fluidly between those layers is likely to be extremely commercially valuable.

We see math as code and code as math. The real magic, and the key transition, comes from combining AI, programming languages, and mathematics—bringing all three pillars together. What we envision is humans using informal reasoning and intuition as a powerful guide, with formal systems then verifying those ideas. That interplay across layers is, I think, the real magic of combining multiple levels of abstraction.

We try to make decisions through a kind of committee-based approach around what species we're going to bring back, where we're going to invest our dollars, and what fits our core ethos. If it's not going to be beneficial to humans, if we can't do it at arm's length from humans, and if we can't do it in a way that's good for animals, then we just don't touch it.

At Colossal, we are not going to work in humans or non-human primates because we felt like we're already going to have an uphill battle with transparency and education, and we don't want people to be like, if a hair-loss treatment comes out of Colossal, 'are they selling a gene from a woolly mammoth?'

From day one, we said that we wanted to create value, create impact, and create inspiration. We're not going to build a business that just creates impact—we're not a nonprofit. We're not going to just create a company that inspires the next generation. We have to create value.

Data is showing that the amount of information we're sending them without hope, and the ratio of hope versus negativity, is shutting down the next generation, which is terrifying. And so it's not like, 'Oh my gosh, there's a fire—they're going to go put the fire out.' They're like, 'Oh my gosh, what's the point? The fire is too big.'