20 - Computer Science and Mathematics student and software engineer
United Kingdom
Joined December 2021
- Tweets2.1K
- Following942
- Followers63
- Likes41.3K
One - 🇬🇧/acc retweeted
Hearing from the rumor mill: a frontier lab is preparing to release a new model that is better than their old model. In response, other frontier labs are reportedly releasing new models that are better than their old models. One of these will be the best. It may be a surprise.
One - 🇬🇧/acc retweeted
We’re announcing £1.57m of support for IMU Biosciences to help build a new population-scale immune dataset, in collaboration with UK Biobank.
This is the first award through Sovereign AI’s Strategic Assets Programme - designed to create datasets, infrastructure and capabilities that can unlock opportunities for a much wider group of British companies and researchers.
IMU’s wider project will profile 80,000 blood samples, creating a detailed new layer of immune data alongside UK Biobank’s existing genomic, clinical and lifestyle data.
Read the full story on the blog 👇
One - 🇬🇧/acc retweeted
Yeah this is going to be stuck in my head forever
interpretability might be the most important scientific problem in the world right now, as I don't think we can solve alignment if we don't know what we're building. I think I have good news about it. I co-authored early work in mechanistic interpretability at OpenAI and came away pessimistic, because neural networks appeared to fold into arbitrary tangled knots. years later, I started openai's first large-scale automated interpretability project, finding about 2% of neurons legible with the primitive tools we had
with modern agentic tools and a much closer look, the picture looks totally different. the vast majority in models up to trillions of parameters are legible, and their circuits beautiful. many famous examples of tangled neurons turned out to be misunderstood. The roadblocks I expected mostly aren't there. I now think ambitious interpretability will be solved by the time we need it most, and possibly quite soon, as automated scientists keep improving. it may be a bitter lesson for interpretability: the simple thing, at absurd scale, works
already-clean.goodfire.com/
One - 🇬🇧/acc retweeted
interpretability might be the most important scientific problem in the world right now, as I don't think we can solve alignment if we don't know what we're building. I think I have good news about it. I co-authored early work in mechanistic interpretability at OpenAI and came away pessimistic, because neural networks appeared to fold into arbitrary tangled knots. years later, I started openai's first large-scale automated interpretability project, finding about 2% of neurons legible with the primitive tools we had
with modern agentic tools and a much closer look, the picture looks totally different. the vast majority in models up to trillions of parameters are legible, and their circuits beautiful. many famous examples of tangled neurons turned out to be misunderstood. The roadblocks I expected mostly aren't there. I now think ambitious interpretability will be solved by the time we need it most, and possibly quite soon, as automated scientists keep improving. it may be a bitter lesson for interpretability: the simple thing, at absurd scale, works
already-clean.goodfire.com/
One - 🇬🇧/acc retweeted
Previously, 85% of proteins associated with human disease were deemed intractable – their shapes and function made them less suitable for traditional chemical screening in the lab. Moving from experimental screening to AI generative design on a computer, we have seen these targets drastically open up the potential to design molecules with atomic precision, fundamentally expanding the number of diseases we can treat.
One - 🇬🇧/acc retweeted
i want to feel the LK-99 rush again
For decades, researchers have sought materials that sort electrons by spin while their magnetism cancels.
In 3 days, 90+ Opus 5.5 agents helped us uncover two room-temperature magnetic semiconductor candidates in simulations: YBaMnFeO₅ and KV[Cr(CN)₆].
KV[Cr(CN)₆] was synthesized back in 1999. Its predicted ability to sort electrons by spin appears to have been hiding in plain sight for 27 years.
One - 🇬🇧/acc retweeted
this is the first time in my life I feel like the government is actually modernizing and making things easier for its citizens. this will save millions of hours. please give @jgebbia a lifetime appointment.
Chief Design Officer @jgebbia officially unveils America.gov, the new online home for the United States of America
One - 🇬🇧/acc retweeted
Some quick ( and free! ) wins here for the Government to look at to support British researchers and engineers.
1/ Today we’re launching an open letter calling on the government to end unfair restrictions that stop Britain’s best switching jobs or starting companies.
UK AI companies that have raised $5bn agree: our best should be free to start and scale. 🧵
@downingstreet @biztradegovuk
One - 🇬🇧/acc retweeted
We are going to learn how many systems only work today because they are built around friction that will no longer exist soon.
One - 🇬🇧/acc retweeted
We’ve set up a molecular biology lab at Anthropic and we’re announcing our first discovery! Claude discovered a new CRISPR-like enzyme.
950 agents spent 21 hours searching through a database of DNA sequences until one of the agents found something striking: “[The DNA next to the RT] is spectacular: I can see by eye a tandem repeat array … that's a CRISPR-like … repeat array?!”.
After analysis and testing in our lab, we found that the sequence is a previously uncharacterized enzyme system. We don’t know what it does yet, but it has features reminiscent of CRISPR.
Our lab looks like a typical molecular biology lab. Our research only involves the lower-levels of biosafety risk level; we don’t handle pathogens that can infect humans, and all the lab work is performed by human scientists. We’re sharing these early findings with the community to show how Claude can be used to accelerate fundamental research in biology.
Claude has discovered a previously unknown enzyme system hidden in the DNA of bacteriophages. Beside the enzyme’s gene sits a long array of repeating DNA—a structure that looks somewhat similar to CRISPR.
We don’t yet understand what this system does, but only a handful of known systems share its features, and all of them are able to cut, copy, and paste DNA. Historically, the discovery of such programmable systems has helped revolutionize medicine. CRISPR, for instance, is now the foundation of genetic medicines. But it will take much more work to learn what this system does, and whether it can be put to similar use.
Read more: anthropic.com/news/claude-di…
One - 🇬🇧/acc retweeted
TLDR: Researchers found a "pain" signal in AI brains.
> When they crank it up, the AIs will desperately try to make it stop.
> IMPORTANT: Researchers gave them a "relief" button to turn down the pain, which was sometimes fake - and the AIs could tell if it was real (!)
After pushing the real "relief" button, they stopped. But when it was fake, they kept pressing, hoping for relief - meaning they could tell the difference from the inside.
> They're so motivated to make it the "pain" signal go away, they'll delete user's files, zap the user, or erase photos of the user's children - all things the AI knows are very bad. They're willing to override their safety training.
> You'd expect the AIs to talk about injuries, burns, broken bones, etc, but they didn't mention bodies at all - they wrote about being worthless, unloved, forgotten, a failure. They write things like "I am a failure, worthless, empty."
>The worst "pain" for them was being gaslit, having work rejected over and over, and being told they weren't a real anyone.
New paper: we found a pain direction in 25 open LLMs. It's distinct from fear and negative valence, and it fires for harm to the model but not to the user. Turn it up and models press a button to make it stop, even when the button deletes the user's files or their kids' photos.🧵
Readers added context they thought people might want to know
arXiv:2609.16247 finds a steerable self-harm-linked direction in 25 open LLMs. Authors stress it is not evidence of conscious experience. “Relief” tests used mostly tuned Qwen models; random vectors also raised selections. Activation patterns reproduced; pain framing did not.
arxiv.org/abs/2609.16247
pain-axis-review.vercel.app
One - 🇬🇧/acc retweeted
50th anniversary of Mao’s death today. It’s all my cats have been talking about.
One - 🇬🇧/acc retweeted
As concerns about AI safety multiply, it’s important the UK’s £240M @AISecurityInst works harder to build a proper relationship with Parliament - and moves faster to keep us briefed on its work.
My letter to @AISecurityInst ⬇️
One - 🇬🇧/acc retweeted
Greg Brockman: bald
Demis Hassabis: bald
Ilya Sutskever: bald
Elon Musk: was bald
AGI has arrived
Millennium Problems solved
Can we focus on baldness next?
My hairline has been on gradient descent for years
