Adam Marblestone
@AdamMarblestone
Technologist, Scientist Co-founder and CEO @Convergent_FROs Tweets do not represent employers
After an awesome two years at DeepMind, I’m leaving tomorrow and will be working full-time on identifying and roadmapping promising new “mini-Manhattan projects”, i.e., “Focused Research Organizations (FROs)”, and trying to help them happen. Talk to me if you have one in mind!
In a new paper for the Day One Project, @SGRodriques & @AdamMarblestone call for the next administration to create “Focused Research Organizations” to tackle scientific and technological challenges & fill a key structural gap in the nation’s R&D system. dayoneproject.org/post/focused-r…
Everyone has heard that Lean is a programming language that allows for proof verification in mathematics. But what does that actually mean and how does it work? If you’re interested in this question you should check out an article my friends and I wrote detailing the nuts and…
Join us live from the birthplace of Silicon Valley @ 3 PM PT to be among the first to experience Mathematical Superintelligence: x.com/i/broadcasts/1…
How to make an object travel quite fast. youtu.be/MDM1COWJ2Hc Not fast enough by a long chalk, but quite fast. We need ideas to come fast too. Much faster than they are coming now.
ashlee & the @corememory team are among the best storytellers in science. we were privileged to host them for a look inside @newlimit. we hope this is only the first of many windows into our mission to add healthy years to each life.
Our new @corememory video on @newlimit, which is finding combinations of proteins that reverse aging across the body. It is perhaps the most exciting work in the bio-tech field. Backed by @brian_armstrong @patrickc @collision @JoshuaKushner @natfriedman and others
Fantastic post. Useful intuitions on both why we discovered the major drugs of the past in the order we did, and why we should be quite hopeful for the next decade.
why don’t we have more medicines? inventing medicines has grown less efficient, limited by our knowledge of biological targets for the 1st time, AI & genomics have the potential to make this knowledge abundant – one of the largest potential impacts of the intelligence age.
it's an exciting week to feel the FROmentum! starting last friday, when the folks at Canon made an exciting post about a rather striking telescopic array pointed at the night sky:
Exciting news: @leanprover, a FRO incubated under Convergent Research, will receive a $5M grant. This forms part of a larger $10M donation, including to Mathlib, to accelerate formal math. More here: renaissancephilanthropy.org/news-and-insig…
What happens when you mix: 🤖 deep learning, ⌚️ new hardware, & 👥 1000's of people? 🧠 The first general neuromotor interface, and a @Nature publication!
We’re thrilled to see our advanced ML models and EMG hardware — that transform neural signals controlling muscles at the wrist into commands that seamlessly drive computer interactions — appearing in the latest edition of @Nature. Read the story: nature.com/articles/s4158… Find…
New blog post: “Behaviorist” RL Reward Functions Lead To Scheming. I argue that, if RL is used to push AI capabilities towards AGI, it will eventually lead to AI that “schemes” (feigns niceness, while secretly looking for opportunities for escape, world takeover, etc.) (1/3)
If the US government wants to lead the world in dataset creation as outlined in the AI Action Plan, open source the @NIST Chemistry WebBook today. Or at least make a purchase for download page.
The neural basis of species-specific defensive behaviour in Peromyscus mice nature.com/articles/s4158…
HLE has recently become the benchmark to beat for frontier agents. We @FutureHouseSF took a closer look at the chem and bio questions and found about 30% of them are likely invalid based on our analysis and third-party PhD evaluations. 1/7
It's FRappening
Today the @WhiteHouse released America’s AI Action Plan to win the global race. We need to OUT-INNOVATE our competitors, BUILD AI & energy infrastructure, & EXPORT American AI around the world. Visit AI.gov
Do you work on alt proteins or in a related space and are you in NYC on July 29? Would you like to help @RenPhil21 and @metaculus try out cool forecasting things? Please DM me! @celloMolly
Turns out they said yes! statnews.com/2025/07/21/cas…
Maybe @Google or @GoogleDeepMind should fund CASP, given their ample resources and how much they've benefited from the efforts of John Moult, Krzysztof Fidelis, Andriy Kryshtafovych, Nick Grishin, and many others. science.org/content/articl…
I give up
lmao crazy result, in case you somehow thought AIs weren't weird enough yet
spicy
Each of our bimonthly HTS data releases reveals previously unknown drug/off-target interactions, but some, like this snippet from last week's Release 5, showing that capsaicin is a highly selective agonist for the orphan nuclear receptor, ESRRB, deserves special note. 🧵 1/4 🧪