Sep 25, 2026 · 1 hr 4 min · 11 segments
00:00 – Intro: AI-Designed Viruses and the Stanford Breakthrough02:10 – The Dual-Use Dilemma & Biosecurity Frontiers04:15 – Recreating Pathogens: Horse Pox & Smallpox Precedents06:25 – Anthropic…
Kevin EsveltGuest
Scott CarneyHost
[gentle music] You know, you were arguably one of the inventors of gene drives, at least in using CRISPR in gene drives.
But once you had created this, this amazing new technology, um, you were pretty worried about what that technology might do.
So can you tell me a little bit about what that work is and what your fears are about your life's work?
And this could cause this gene drive effect that drives the change through a whole population, which is great if you wanna change all the mosquitoes so they can't spread malaria or other diseases.
But the next day I woke up and thought, "Wait a minute, wait a minute, wait a minute.
Just who's gonna be able to do this?" Because if you engineer one critter in the lab and you let it go, then there's a chance it'll die before it mates.
If you can engineer a mosquito to never spread disease, you can engineer a mosquito to spread disease with every bite.
Is this a problem? Could this be a weapon? Because anything that can spread on its own autonomously, making more and more and more copies of itself, seems pretty dangerous.
So I wanted to figure out how do you tell whether your technology is net good or bad? Is it offense-
the classic example of rabbits being introduced in Australia, right? And then now rabbits are everywhere.
Like, "Oh my God, what are we gonna do with all these rabbits?" But you're like, "No, we're gonna create a mosquito that, you know, will never, uh, be able to reproduce." But maybe someone else could put a mosquito out that every time it bites you, you get alpha-gal syndrome, and now you're allergic to meat or whatever, right?
And how can you even imagine how you would stop something from being bad? Like, it seems like s- it's sort of like that, that Jurassic Park, uh, thing, you know, where, where there's like, "Well, just because we could do it doesn't mean that we should."
Once you put it out there, other people can use it, and you don't have absolute control over what they're gonna do with it.
So your most powerful lever is deciding whether or not it's safe to tell the world about that thing in the first place.
[gentle music] You know, you were arguably one of the inventors of gene drives, at least in using CRISPR in gene drives.
But once you had created this, this amazing new technology, um, you were pretty worried about what that technology might do.
So can you tell me a little bit about what that work is and what your fears are about your life's work?
And this could cause this gene drive effect that drives the change through a whole population, which is great if you wanna change all the mosquitoes so they can't spread malaria or other diseases.
But the next day I woke up and thought, "Wait a minute, wait a minute, wait a minute.
Just who's gonna be able to do this?" Because if you engineer one critter in the lab and you let it go, then there's a chance it'll die before it mates.
If you can engineer a mosquito to never spread disease, you can engineer a mosquito to spread disease with every bite.
Is this a problem? Could this be a weapon? Because anything that can spread on its own autonomously, making more and more and more copies of itself, seems pretty dangerous.
So I wanted to figure out how do you tell whether your technology is net good or bad? Is it offense-
the classic example of rabbits being introduced in Australia, right? And then now rabbits are everywhere.
Like, "Oh my God, what are we gonna do with all these rabbits?" But you're like, "No, we're gonna create a mosquito that, you know, will never, uh, be able to reproduce." But maybe someone else could put a mosquito out that every time it bites you, you get alpha-gal syndrome, and now you're allergic to meat or whatever, right?
And how can you even imagine how you would stop something from being bad? Like, it seems like s- it's sort of like that, that Jurassic Park, uh, thing, you know, where, where there's like, "Well, just because we could do it doesn't mean that we should."
Once you put it out there, other people can use it, and you don't have absolute control over what they're gonna do with it.
So your most powerful lever is deciding whether or not it's safe to tell the world about that thing in the first place.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.