Scientist against lab-based pandemics 🧬 Co-author of VIRAL: the search for the origin of Covid-19 📖 A dangerous young investigator 🕵🏻♀
Cambridge, MA
Joined December 2011
- Tweets25K
- Following607
- Followers78.5K
- Likes16.4K
Pinned Tweet
How to Stop a Lab Leak from Starting the Next Pandemic.
My new op-ed made possible by the @nytimes graphics and editorial team:
nytimes.com/interactive/2026…
Whatever one believes about AI-facilitated outbreaks (lab leaks or bioterror), it makes sense to have an independent agency track the sale of enabling materials (pathogens, sequences, enzymes, cell lines, production & purification reagents) even when selling to established labs.
Contingency plans for the day after an AI-assisted or AI-initiated biological warning shot.
Colman-Lerner's and my "Break-Glass" paper is up.
preprints.org/manuscript/202…
The @BrookingsInst is hosting a fireside chat on Nov 5 (in person and virtual) to assess the July 2026 Dangerous Gain-of-Function policy and legislative proposals such as the Gene Synthesis Screening bill and Risky Research Review Act.
brookings.edu/events/us-bios…
Alina Chan retweeted
The more immediate #AI #bioweapons concern is not that AI suddenly enables amateurs but that it amplifies already capable actors, I told @guardian. The key question is which barriers AI lowers, for whom, and at what points in the biological threat pathway. theguardian.com/technology/2…
A lot has been said about AI x bio trigger events that should lead to increased regulation or oversight.
Yet, ~7 years since Covid emerged, the research pipeline that likely caused the pandemic is still legal to conduct outside of high containment without independent oversight.
A US-based investigation of Covid origins should at minimum:
📧 Access all communications and data shared between the Wuhan Institute of Virology and US/international partners including but not limited to the EcoHealth Alliance (now shut down) and University of North Carolina
🧪 Assemble a team of scientists to curate and re-analyze relevant data and methodology e.g. relating to early cases, SC2 and close relative sequences, market and other animal/human sampling efforts
🕵️♀️ Evaluate the full range of intel available to US and its allies e.g. Germany's BND assessed lab origin with 80-95% certainty, US doctors in Wuhan said they heard of novel respiratory disease in mid Dec 2019, US intel says military was engaged in coronavirus research at the WIV, pangolin virus samples were sent there, and WIV staff were ill with possibly Covid in Nov 2019
📒 Search funding agencies, journals, databases and even commercial suppliers in the US and its allies for relevant proposals, reports, manuscripts, data entries, purchases
And ideally be bipartisan and even international 🌏 in nature, appointing investigators and scientists with a wide range of starting points, so that its findings are not easily politicized.
All of this could've been started in 2020 but here we are almost 7 years later with no such investigation in sight for a pandemic that harmed billions of people, killed at least 15 million people, caused an estimated $16 trillion in economic loss to the US alone, and pushed tens of millions of people around the world back into poverty and debt.
Without a bipartisan or international consensus on how the pandemic most likely happened, little has been done to prevent it happening all over again. If Covid can be considered a "trigger event" to warn the world of worse bio-catastrophes, there hasn't been any meaningful course correction.
No punches pulled by @bgurley @theallinpod on the still unresolved Covid origins, calling out influential virology insiders who were blockers of an investigation.
"My p(2nd Covid) is way higher than my p(doom). And if their models are so ****ing smart, tell me what happened here."
All-In Summit: Bill Gurley
(0:00) Bill Gurley takes the stage at the All-In Summit
(1:29) Bill's Presentation: Searching for Feynman
Links to additional resources mentioned in talk and used in research:
p3.institute/searching_for_f…
@bgurley
Almost 7 years since Covid emerged, there's still no US strategy for tracking and governing the type of research that likely caused the pandemic.
The closest thing to new legislation has not been scheduled for a Senate vote in over a year.
congress.gov/bill/119th-cong…
The most plausible AI-driven pandemic scenario in the next 10 years is a well resourced, potentially gov/state-funded group exploring AI pathogen design for biodefense/pandemic preparedness, followed by a lab leak.
Not a random lovesick teenager ordering AI bioweapons online.
Important detail about Anthropic's case studies of biological misuse:
"The individuals implicated in these case studies are working scientists. We do not assert that they intended harm..."
Possible these scientists were testing the potential for misuse with good intentions.
I am really not surprised that the first people seriously testing AI's ability to enhance pathogens and toxins are scientists and that they work for governments with significant biodefense programs.
A related discussion is about well-resourced actors creating pandemic pathogens with the help of AI. @Noahpinion links to a detailed discussion of this by @owl_posting who wrote:
“I am consistently on the side that only a well-funded actor would be capable of such a thing. And why wouldn’t those actors opt for much simpler acts of violence that would roughly accomplish the same thing?”
My answer to this is: professional advancement.
There are scientists whose promotions and funding are dependent on them creating pandemic pathogens - sometimes even with good intentions like "we're making the AI biothreats so we can create countermeasures for them". This is a much more terrifying and realistic scenario than the rejected teen.
owlposting.com/p/reasons-to-…
