Jack Szostak – Life’s Next Transition May Not Be Biological (Stewarding the Flame, Episode 10)

ByAnnelize Potgieter November 6, 2026November 6, 2026 November 6, 2026

This new installment of the Stewarding the Flame series is an interview with Jack Szostak, University Professor in the Department of Chemistry at the University of Chicago and recipient of the 2009 Nobel Prize in Physiology or Medicine for his work on telomeres.

What distinguishes Jack’s approach is that, for all his decades of work on how life gets started, he is not married to any particular biological version of how life must function. He is reluctant to define life, because the word is used for so many different things. What he works on is the transition from chemistry to biology. For him, the fundamental aspect of biology is evolution: the origin of life is the question of how simple chemistry gets to a system that can do Darwinian evolution. Whether that can happen through one set of molecules or many, he says, we don’t yet know.

In this conversation, Jack discusses possible steps and open questions on the path from heritable information to cells and multicellular life. He explains why he sees no overarching conceptual barriers to understanding how life got started, and why he doubts consciousness would spontaneously emerge from large neural networks. He also discusses how self-improving AI might eventually lead to new types of physically instantiated minds.

The interview is the tenth installment in The Trajectory’s Stewarding the Flame series, where we ask: What is intelligence, and what is the “flame” that life has, which non-life does not?

This series references the article: Stewarding the Flame – How to Build an Ideal Future for Intelligence.

I hope you enjoy this fascinating conversation with Jack Szostak:

Subscribe for the latest episodes of The Trajectory:

Below, we’ll explore the core takeaways from my conversation with Jack, focusing on three of the central questions that shape the Stewarding the Flame series: what makes life, life, what humanity still needs to discover in order to better understand the living process, and how humanity can contribute to carrying that process forward.

What Makes Life, Life?

Jack doesn’t like trying to define life, because the word is used to talk about many very different things. His own concern is the transition from chemistry to biology, so when he talks about life, he means biological life. Even within biology, he notes, there are plenty of edge cases, from things we call alive that don’t reproduce to viruses. Life in silico, in the context of AGI, is a different thing again.

For Jack, the fundamental aspect of biology is evolution. The origin of life is really the question of how to get to a system that can start to do Darwinian evolution. That requires a way of encoding information: a genome. His lab is working on a model of what the first genome might look like, involving autocatalytic sets of small RNA molecules. As evolution proceeds, the genome accumulates information about the environment and about its own cellular structure. With an open-ended way of encoding information, that system can just keep going.

The closer we get to building these systems, the less there seems to be a single line between life and non-life. Low-level evolution can begin with simple enrichment for sequences that are good at being copied. Then there is a step change, when sequences start to encode a useful function. As functions multiply, the system becomes something more and more people would recognize as life, but different people will draw the line in different places.

Jack considers one of the most interesting questions about the origin of life to be how many ways it could have happened. Some researchers increasingly think the inherent chemistry may focus life toward one set of initial molecules. Others think that if there’s one way to get life started, there must be others, perhaps ten, perhaps a million. Exploring other origin scenarios and other chemistries that could lead to evolving systems is, in his words, really challenging, and a huge challenge to the field.

“At this point we don’t know. We have a lot of knowledge and understanding of one way, which is what led to us.”

What Do We Still Need to Discover to Understand the Flame?

Jack is confident that minds like ours are capable of understanding how life began. Fifty years ago, researchers didn’t know that RNA could be an enzyme, and the techniques were so primitive that many people concluded life might be too complicated ever to understand. Progress in recent decades has given us a pretty good overall model of how life got started. The work now is going into more and more detail on the individual steps.

He expects a major challenge of the coming decade or two to be the path from the first simple system with heritable information to a few functions. From there, a progressively larger genome can encode more functions, likely first in RNA enzymes, then more complicated metabolism, and eventually protein synthesis up to LUCA, the last universal common ancestor. What it takes to become multicellular is another open question. Jack points to archaea that appear to have cytoskeleton components and may have been amoeboid, the engulfed bacteria that became mitochondria and chloroplasts, rising oxygen, the poorly understood Ediacaran fossils, and the Cambrian explosion.

Jack warns that many biologists treat Darwinian evolution as so powerful it can do anything, and forget that it operates within the laws of chemistry and physics. For roughly two billion years Earth was basically a slime planet. Nothing very interesting happened until it became possible to accumulate higher levels of oxygen. For all the big transitions, he would like to understand why they happened when they did, why they didn’t happen earlier, and how much of our history depended on random, stochastic events, like the impact that wiped out the dinosaurs.

He sees the evolution of nervous systems as a frontier. Perception of the environment starts early, with bacteria sensing their external and internal environments. Those networks of stimuli and responses become more complex over evolutionary time. For anyone working on computer consciousness or AGI, he argues, the evolution of consciousness and values in animals is the one example we have, and we need to understand it much better. Like life, he doesn’t think consciousness is binary.

What should and shouldn’t we do with AGI and BCI if we want the best chance for the greater process-of-life to flourish?

Jack suspects today’s AI systems, which you can talk to like another person, have no self-awareness or sentient experience. He doubts it would spontaneously emerge from an extremely large neural network trained with gradient descent, because in animals these states took a long time to evolve. But he sees no reason it couldn’t be gradually designed in. He expects that AI systems running around as humanoid robots will need information about their internal state to work better: first their bodily state, and at some point their internal mental state. He points to hallucination as an example: how do you stop a system from making things up if it isn’t aware that it’s doing it?

He doesn’t hold a strong intuition of his own about life manifesting in non-biological systems. In Is Earth Exceptional?, it was his co-author Mario Livio who speculated that life in the universe might be mostly not biology, with machines spreading through the universe where we cannot. Jack calls it possible. Where he sees a path is in speed: cultural evolution goes many orders of magnitude faster than Darwinian evolution. AI systems are now good at writing code, and he suggests they should eventually be able to improve their own capabilities. If they can also redesign their internal structure, he says, that may be a path to consciousness and everything else. By analogy with our new ability to redesign our own structures, Jack can easily imagine sufficiently advanced AI becoming very good at hardware design, leading eventually to “a whole diversification of new types of physically instantiated minds.” He adds that we are manipulating our own genomes too, and calls this urgent to think about.

Jack notes that every major evolutionary transition brings a blossoming of new forms of life, while most things that were there before die out. Oxygen, for example, is incredibly toxic to organisms that haven’t evolved to live with it. Jack would personally rather not see humanity wiped out by an age of machines. He doesn’t think we have a lot of control over the future of this process, and he is clear that stasis is not an option. Humans won’t look the same in a million years. If intelligences with emotions and human-level complexity succeeded us, accepting that as a long-term transition would take, he says, being “a little bit selfless.”

Asked what would help the process flourish, he goes first to the baseline: adequate investment in the sciences. He argues that the current political situation in the United States is damaging its scientific infrastructure, by discouraging students from academic tracks and reducing funding for grad students and postdocs. Other places in the world prioritize science, and if the US isn’t in the lead, they will be. He also values the multidisciplinary environment of a real university. Alongside physics, chemistry, biology and neuroscience, it holds the humanities, economics and law, so that people aren’t only thinking about the next technical advancement.

Concluding Notes

Throughout this conversation, Jack presents life as the outcome of a step-by-step path from chemistry to Darwinian evolution. It is not a single threshold, and the line between life and non-life gets harder to draw the closer we get to building it. He treats consciousness the same way: a gradual development across animal lineages, not a binary switch. He believes that understanding how life began is within our reach.

What stood out most to me was that Jack, in parallel, came to the same conclusion as a relatively eminent Google DeepMind researcher at one of our physical events. In non-biological systems, self-transcendence and the evolutionary process could start as recursive self-improvement in an entirely digital environment. It could then extend to robots, and to robots building additional robots.

Much of today’s biosphere consists of entities that develop: they start small, and their shape unfolds over time. Systems that write their own code and construct themselves out of parts could undergo Darwinian evolution through mechanisms totally separate from how today’s biolife does it. That suggests many mechanisms we consider sacred to life may not be required at all. Jack was wildly open-ended about this, and I’ll be doing more journaling, and potentially some articles, on it.

These are not waters Jack normally swims in, but he jumped right in and added a perspective we haven’t had on the show before. His call to fund fundamental science and to overtly encourage multidisciplinary thinking is very much in line with what we’re trying to do here.

I hope you’ll join me for the next conversation in the Stewarding the Flame series as we continue exploring what intelligence is, what makes life, life, and what it will take to carry the greater process of life successfully into the future.

Follow The Trajectory

• Sign Up for The Trajectory Newsletter
• Subscribe to The Trajectory on YouTube
• Subscribe to The Trajectory podcast
• Follow on LinkedIn
• Follow on X