60: Patrick Hsu - Speedrunning the Infinite Game

Patrick Hsu (0:00)

When I was 13, I got expelled from college.

Jackson Dahl (0:01)

Expelled from college? You were already in college?

Patrick Hsu (0:04)

Well, not anymore. Research is 99% failure. All your very best ideas kind of suck, except the one or two that you're remembered for. Research is an even playing field, by definition. No matter how great or accomplished you are, we all have the same shot. I would say research taste and intuition is only increasing in relative importance. Like I would buy call options on it, if I could.

Jackson Dahl (0:24)

Getting to the point where a model that makes a cell biologist feel emotion, can you say more about what that means?

Patrick Hsu (0:29)

You can take your own fibroblasts and turn them into induced pluripotent stem cells in a dish. It is one of the most profound and magical and semi-religious feelings I've ever had. How do we have those holy shit moments in biology?

Jackson Dahl (0:42)

Do you have many of those currently?

Patrick Hsu (0:43)

Well, we've created them and the first time that AI has been used to create entire viable organisms, I think we will create evolutionarily distinct forms of life. A hundred years ago, infectious disease was maybe the number one killer. Today, that's heart disease. We may see both of those go away. One of the big ironies of writing papers is, I don't really know if I'm writing them for fellow humans anymore.

Jackson Dahl (1:08)

Do you fear becoming obsolete?

Patrick Hsu (1:09)

If you simply do nothing, the default is to become obsolete. So it's not even a fear, it's actually just reality. Being able to be bold, to be fearless, to be risk-taking, they're survival traits. I think science is really hard. It's such a long journey and it's mostly failure. So if you don't have a way of bringing the energy and the vibe and the vitality, it's going to be pretty miserable. You need to figure out a way to make the discovery process fun because you're mostly eating shit.

Jackson Dahl (1:39)

Welcome to Dialectic, episode 60 with Patrick Hsu.

Patrick is a scientist, a founder, an investor. He co-founded the Arc Institute and is a core investigator there. Arc is a new type of research organization in collaboration with a number of universities that is doing groundbreaking work at the intersection of technology and science. Patrick is also an assistant professor of pathology at Stanford University and a venture partner at Thrive Capital. Patrick works at the rare intersection of biology, engineering, and artificial intelligence, and has worked across a number of different innovations over the course of his research career, including CRISPR very early on, and most recently on bio foundation models, including Evo 1 and Evo 2, as well as cell models. He is a true polymath, and it was so fun to sit down with him and talk about Arc, all of the work they're doing to use AI to push the frontiers of biotech, what might be possible, what makes these institutions work, and more.

Patrick has also been incredibly prodigious going to college as a young teenager and has continued to find new ways to grow, reinvent himself, and continue to push what is possible in the domains that interest him. And frankly, those domains keep increasing. I hope you are inspired to raise your bar, continue to reinvent yourself, and to have fun in the process. If you enjoy the conversation, please share it with a friend. I'd also like to thank Notion, Dialectic's presenting partner. Patrick is actually on the board of Notion, and Notion is a collaborative workspace for your life's work. It is a tool you can use alone or with your collaborators to create documents and databases, collaborate across systems, and get leverage from the increasingly powerful AI and agent tools that so many people are so excited about. The great thing about Notion is that it plays a Switzerland role. It is friendly with every AI in that it integrates with whatever model, whatever tool, whatever agent you're using. Not only that, but models are available typically in Notion the day they release and you can tag any agent from within Notion right in the document, or tag Notion from within whatever other AI tool you might be using. You can even connect Notion to agents like Muse or Instinct and tell them to fill all the context, memory or important information from your conversations into Notion so that you have that context wherever you go. Even as all of the tools around you might change as capabilities are changing, you can trust Notion as the single place that doesn't change and it will work with whatever tool you use next. You can learn more about Notion or check it out if you haven't in a while at notion.com.