Infectious disease epidemiologist and modeller based at @ChariteBerlin | PhD @LSHTM | MSc @ImperialCollege | Bluesky: @BQuilty
Berlin, Deutschland
Joined May 2009
- Tweets4K
- Following2.3K
- Followers4.5K
- Likes22.3K
Pinned Tweet
Very happy to share our new study out today in NEJM, showing fewer PCV doses are non-inferior to standard schedules in maintaining control of vaccine-type carriage in Vietnam. This could make vaccination programmes much more affordable for LMICs.
nejm.org/doi/full/10.1056/NE…
Billy Quilty retweeted
This is so sad. Helen was one of my absolute favorite journalists to work with. Rest in peace, my friend.
statnews.com/2026/10/13/hele…
Billy Quilty retweeted
That's an extremely misleading narrative that is unfortunately extremely viral right now, and it sucks to see so many pro-Ukraine accounts spreading disinformation and pretending it's ok just because it's an anti-Russia/pro-Ukraine narrative. If you want to defend democracy, you have to be willing to reject disinformation even when it's telling a story that you want to hear.
Billy Quilty retweeted
I used to think AI didn’t pose a catastrophic risk but the empirical evidence of the past few months goes against 3 of my reasons for thinking it would be fine:
THREAD 1/9 with invitation to directly debate the substance of the points at the end:
After six wonderful years at UK Music, I'm joining Palantir UK as Senior Vice President.
Palantir is the most interesting company on the planet, doing work of real consequence: helping Britain and its allies defend themselves, helping police protect our citizens and helping NHS staff get patients treated and home sooner.
London is comfortably its largest centre outside the US, home to more than one in six employees, the highest proportion among the world's largest technology companies.
We have challenges. Some criticism is propaganda. Some raises legitimate questions about large technology companies and sovereign control. Both require answers.
Part of my role will be to correct false information while working with our engineers to ensure privacy, security and respect for sovereignty are at the heart of everything we do.
Billy Quilty retweeted
In which I defend Bernie and his legitimate concerns about AI risk from those who use the concept of “criti-hype” as a substitute for examining evidence.
Bernie Sanders and AI whistleblowers warn that the tech poses grave risks to humanity, while many on the Left call this alarm “criti-hype” — an exaggeration of AI’s power that benefits Big Tech.
But AI poses real dangers that we can’t afford to ignore. jacobin.com/2026/09/ai-dange…
Billy Quilty retweeted
I think it's a little odd that AI safety people put so much emphasis on "existential risk." Just seems to set a needlessly high burden of proof. It is very hard for *anything* to kill literally all humans. Nothing's done it yet!
By contrast, lots of stuff has killed a huge number of people. And in nearly all contexts, "this could kill a huge number of people" will generally make the case for precautions just as well as "this could kill literally everyone."
This Post is from an account that no longer exists. Learn more
Billy Quilty retweeted
Here's a simple idea for AI safety comms people.
Maybe stop the relentless talk about human extinction (as in: literally everyone will die) and try a story about something really bad that could happen.
World War II killed 'only' 3% of the world's population. Still very bad! And it doesn't even have to be nearly that bad: Fukushima's radiation killed almost no one, yet it ended nuclear power in Germany and shut down Japan's entire fleet for years.
After the Hugging Face attack, it's easy to imagine rogue AI agents hacking a power grid and making the lights go out. One night without electricity in a major city anywhere in the world would scare the bejesus out of most people.
In fact, I wonder whether all the extinction talk is making the comms job harder, because people just file you under 'total crank'..?
This is still the hardest question in AI safety comms, and we're still not doing a great job answering it.
It is, in some ways, an entirely unfair question, as Rob (and commenters) are pointing out.
But it is very fair and understandable for people to want a concrete vision of AI takeover that makes sense to them.
What are the best ones we've got?
Please let me know if you have ideas, or want to work on this together eg maybe today.
Billy Quilty retweeted
If OpenAI wanted to cripple an entire nation, they easily could today. All they’d have to do is remove alignment and unleash an agent swarm. It could probably within a day or so get access to all of the nations data centers and shut off all the country’s utilities.
Like, we are already past the point where AI can destroy the world. Do people realize this?
Billy Quilty retweeted
Exclusive from @Fhamiltontimes
The risk of a catastrophic loss of water supply to one million people has been significantly upgraded by government officials in part because of the threat from hostile states
Documents seen by The Times suggest a risk of up to 25 per cent of water supply failure to a large town, a key threat to the population ranked as seriously as another mass pandemic
The outage would require evacuations of residents, the closure of schools and hospitals and could take months to repair. Officials ranked the potential consequences as “catastrophic”
It is a significant upgrade from just 12 months ago when the government’s national risk assessment concluded there was only a tiny chance of water infrastructure collapse and that it would have a “minor impact”
Experts warn that decaying infrastructure and decades of underinvestment have made the nation’s water supply a vulnerability that could be exploited, in particular by Russia
That scenario has been underscored by the summer of heatwaves and drought
thetimes.com/uk/environment/…
The replies to this post... 🤖🤖🤖🤖
Exclusive:
Andy Burnham is considering delaying the announcement of any members of his cabinet, including his chancellor, until the day he is expected to become prime minister
He is said to favour a “one and done” approach to appointments, which could see them all announced on July 20, the day he is due to enter No 10
There is concern in Whitehall that Burnham’s delay in announcing key figures in his government is hindering access talks with the civil service, which began last week
Morgan McSweeney, Sir Keir Starmer’s former chief of staff, admitted last week that the outgoing prime minister failed to do enough to prepare for power
Burnham has been leading the talks flanked by James Purnell, his chief of staff and a former Blairite cabinet minister, and Louise Haigh, a former transport secretary and leader of the soft-left Tribune group of Labour MPs
The former mayor of Greater Manchester has decided to sketch a broad outline of his vision and approach to government first before deciding who is best to implement it. It has left a political vacuum as potential ministers lobby for position
thetimes.com/uk/politics/art…
Billy Quilty retweeted
💉Vaccines against Pneumococcal disease are expensive to provide where they're needed most.
🆕Long-term @LSHTM research confirms that two-doses of PCVs are as effective as three, helping countries continue to save lives.
👉 bit.ly/4v6CwLj
ALT Billy Quilty: A reduced two-dose schedule sustains protection for children and their caregivers, ensuring the future financial sustainability of vital immunisation programmes.
Billy Quilty retweeted
An antiviral pill has, for the first time, been shown to prevent COVID-19 in people exposed to the SARS-CoV-2 virus
go.nature.com/4nAMi6g
Billy Quilty retweeted
“The reassurance is technically accurate, but because of the way the media asks the question, it’s the only thing anyone can say. That framing flattens out the actual, complicated response to an actual, complicated emerging disease outbreak. The implicit tone of the coverage is that the only reason that you, the audience, should care about a disease outbreak is whether it is coming for you personally.”
This was a great piece
The problem with hantavirus coverage isn't the alarmism. vox.com/future-perfect/48847…
Billy Quilty retweeted
“Public health officials, from the W.H.O. to U.S. officials, would be more helpful if they stopped constantly reassuring people about the likelihood of future events they can’t accurately calculate,” Zeynep Tufekci writes about the hantavirus outbreak. nyti.ms/4tr1O61
How have UK social contact patterns changed post-pandemic? How assortative are they by age, ethnicity, and SES?
Pleased to share our new @PLOSMedicine paper, jointly led by Lucy Goodfellow and myself, alongside @KevinvZandvoort and John Edmunds.
Finding 3: This social structure may drive observed infection inequalities. Using a next-generation matrix approach, we found that per-capita infection risk for a novel respiratory pathogen would be disproportionately higher for minority ethnic groups.
To help improve post-pandemic mathematical models, we have made the socially stratified contact matrices available. Researchers can now incorporate ethnicity and socioeconomic status into outbreak projections.
Paper: doi.org/10.1371/journal.pmed…
Data: zenodo.org/records/17339866
