Andrew White 🐦⬛
@andrew.diffuse.one
Head of Sci/cofounder at futurehouse.org. Prof of chem eng at UofR (on sabbatical). Automating science with AI and robots in biology. Corvid enthusiast
Making "AI Scientists" has become a hot topic lately. The first reference I could find was from 2008. The term has been used for 20 years! Like "Adam," an AI Scientist robot for studying yeast was published in 2009. I wrote a short post about the term and what it means now.
diffuse.one/p/w1-001
diffuse.one/p/w1-001
October 29, 2025 at 6:26 PM
Making "AI Scientists" has become a hot topic lately. The first reference I could find was from 2008. The term has been used for 20 years! Like "Adam," an AI Scientist robot for studying yeast was published in 2009. I wrote a short post about the term and what it means now.
diffuse.one/p/w1-001
diffuse.one/p/w1-001
It sounds insane, but remember there are 10^14 atoms in a human cell and 10^20 femtoseconds in a day. And across multiple simulation engines, it requires 10^4 FLOPs per atom x femtosecond 2/3
September 26, 2025 at 3:19 PM
It sounds insane, but remember there are 10^14 atoms in a human cell and 10^20 femtoseconds in a day. And across multiple simulation engines, it requires 10^4 FLOPs per atom x femtosecond 2/3
I finished my estimate on required compute to make an atomic-resolution virtual cell: 10^38 FLOPs to simulate a human cell for 1 day. We should be able to do this simulation in 2074 using 200 TW of power. 1/3
September 26, 2025 at 3:19 PM
I finished my estimate on required compute to make an atomic-resolution virtual cell: 10^38 FLOPs to simulate a human cell for 1 day. We should be able to do this simulation in 2074 using 200 TW of power. 1/3
Our ether0 paper was accepted at NeurIPS 2025! Very proud of the FutureHouse team!
September 19, 2025 at 3:42 PM
Our ether0 paper was accepted at NeurIPS 2025! Very proud of the FutureHouse team!
You can also look at it over time. Here's relatively popularity of different animal models in research over time.
Anyway, found this to be interesting. More details about it here: diffuse.one/p/d2-003 3/3
Anyway, found this to be interesting. More details about it here: diffuse.one/p/d2-003 3/3
September 14, 2025 at 4:52 PM
You can also look at it over time. Here's relatively popularity of different animal models in research over time.
Anyway, found this to be interesting. More details about it here: diffuse.one/p/d2-003 3/3
Anyway, found this to be interesting. More details about it here: diffuse.one/p/d2-003 3/3
Here's one measuring the frequency of sample sizes. Like how often people use 8 samples vs 12 samples for reporting research results. N=2 is apparently the most popular 2/3
September 14, 2025 at 4:52 PM
Here's one measuring the frequency of sample sizes. Like how often people use 8 samples vs 12 samples for reporting research results. N=2 is apparently the most popular 2/3
Google scholar has a full-text index of nearly all research papers. You can use it to get counts for arbitrary phrases. I've been using this to measure popularity of things in science. For example, here's the popularity of Greek letters used in equations 1/3
September 14, 2025 at 4:52 PM
Google scholar has a full-text index of nearly all research papers. You can use it to get counts for arbitrary phrases. I've been using this to measure popularity of things in science. For example, here's the popularity of Greek letters used in equations 1/3
I've written up some thoughts on publishing for machines. 10M research papers are published per year and there are 227M total - machines will be primary producers and readers of publications going forward. Humans can simply not keep up. It's time to think about revising the scientific paper.
August 15, 2025 at 6:10 PM
I've written up some thoughts on publishing for machines. 10M research papers are published per year and there are 227M total - machines will be primary producers and readers of publications going forward. Humans can simply not keep up. It's time to think about revising the scientific paper.
It’s a clever question. But it’s not really about frontier science. Multiple papers have shown that Oganesson is not a gas (it’s predicted to be semiconducting solid), it’s not noble (it’s reactive), and it isn’t included in any "terrestrial matter" tables of noble gases. 3/7
July 23, 2025 at 4:29 PM
It’s a clever question. But it’s not really about frontier science. Multiple papers have shown that Oganesson is not a gas (it’s predicted to be semiconducting solid), it’s not noble (it’s reactive), and it isn’t included in any "terrestrial matter" tables of noble gases. 3/7
The design process of HLE required the questions to be unanswerable by contemporary LLMs. That lead to many gotcha style questions like the one below. It’s a trick question – in 2002, a few atoms of a group 18 element Oganesson were made for a few milliseconds. 2/7
July 23, 2025 at 4:29 PM
The design process of HLE required the questions to be unanswerable by contemporary LLMs. That lead to many gotcha style questions like the one below. It’s a trick question – in 2002, a few atoms of a group 18 element Oganesson were made for a few milliseconds. 2/7
I have written up a 3.5k word/10 figure essay on how to write a reward function while avoiding reward hacking for chemistry. It covers all the ridiculous ways we had to avoid reward hacking for training ether0, our scientific reasoning model.
diffuse.one/p/m1-000
diffuse.one/p/m1-000
June 22, 2025 at 3:21 PM
I have written up a 3.5k word/10 figure essay on how to write a reward function while avoiding reward hacking for chemistry. It covers all the ridiculous ways we had to avoid reward hacking for training ether0, our scientific reasoning model.
diffuse.one/p/m1-000
diffuse.one/p/m1-000
The code for this is really minimal - similar to Google Co-Scientist we used multiple agents (from our platform in this case) and tournament-style rankings to select ideas. We're open sourcing it next week, along with all the trajectories.
May 20, 2025 at 3:35 PM
The code for this is really minimal - similar to Google Co-Scientist we used multiple agents (from our platform in this case) and tournament-style rankings to select ideas. We're open sourcing it next week, along with all the trajectories.
The figures, hypothesis, original and follow-up experiments were all generated from our agents. Interestingly, only the lab-work and the paper writing were not automated (which is the opposite of what I would have predicted 2 years ago).
May 20, 2025 at 3:35 PM
The figures, hypothesis, original and follow-up experiments were all generated from our agents. Interestingly, only the lab-work and the paper writing were not automated (which is the opposite of what I would have predicted 2 years ago).
FutureHouse's goal has been to automate scientific discovery. Now we used our agents to make a genuine discovery – a potential new treatment for one kind of blindness (dAMD). We had multiple cycles of hypotheses, experiments, and data analysis – including identify the mechanism.
May 20, 2025 at 3:35 PM
FutureHouse's goal has been to automate scientific discovery. Now we used our agents to make a genuine discovery – a potential new treatment for one kind of blindness (dAMD). We had multiple cycles of hypotheses, experiments, and data analysis – including identify the mechanism.
We shipped multi-agents today! Our chemistry design agent can now call Crow, our scholarly research agents, to bring in data from literature/clinical trials/open targets while designing molecules.
platform.futurehouse.org
platform.futurehouse.org
May 13, 2025 at 3:45 PM
We shipped multi-agents today! Our chemistry design agent can now call Crow, our scholarly research agents, to bring in data from literature/clinical trials/open targets while designing molecules.
platform.futurehouse.org
platform.futurehouse.org
Integrating @opentargets.org is so helpful to provide evidence for disease mechanisms independent of the literature. Here's a demo of synthesizing 78 papers and open targets to propose two novel targets for triple negative breast cancer
See the answer: platform.futurehouse.org/trajectories...
See the answer: platform.futurehouse.org/trajectories...
May 11, 2025 at 1:59 AM
Integrating @opentargets.org is so helpful to provide evidence for disease mechanisms independent of the literature. Here's a demo of synthesizing 78 papers and open targets to propose two novel targets for triple negative breast cancer
See the answer: platform.futurehouse.org/trajectories...
See the answer: platform.futurehouse.org/trajectories...
We have an API for clinical trials on our platform - which means you can ask questions like "what trials will read out in June for NSCLC and how likely would you rate their success based on previous trials in the area." Pretty cool.
Answer: platform.futurehouse.org/trajectories...
Answer: platform.futurehouse.org/trajectories...
May 9, 2025 at 4:54 PM
We have an API for clinical trials on our platform - which means you can ask questions like "what trials will read out in June for NSCLC and how likely would you rate their success based on previous trials in the area." Pretty cool.
Answer: platform.futurehouse.org/trajectories...
Answer: platform.futurehouse.org/trajectories...
Here's a command that converts a DOI to bibtex:
May 6, 2025 at 10:45 PM
Here's a command that converts a DOI to bibtex:
The plan at FutureHouse has been to build scientific agents for discoveries. We’ve spent the last year researching the best way to make agents. We’ve made a ton of progress and now we’ve engineered them to be used at scale, by anyone. Free and on API.
May 1, 2025 at 4:16 PM
The plan at FutureHouse has been to build scientific agents for discoveries. We’ve spent the last year researching the best way to make agents. We’ve made a ton of progress and now we’ve engineered them to be used at scale, by anyone. Free and on API.
And if you want all the functional groups:
I would actually love to have someone explain what the correct answer for this molecule.
I would actually love to have someone explain what the correct answer for this molecule.
March 8, 2025 at 11:18 PM
And if you want all the functional groups:
I would actually love to have someone explain what the correct answer for this molecule.
I would actually love to have someone explain what the correct answer for this molecule.
It's ridiculous, but there hasn't existed a one-liner to quickly get functional groups of a molecule. Little Friday night coding exercise to get this working.
Enjoy - and let me know of any missing functional groups! I could only do a few hundred.
Enjoy - and let me know of any missing functional groups! I could only do a few hundred.
March 8, 2025 at 7:58 AM
It's ridiculous, but there hasn't existed a one-liner to quickly get functional groups of a molecule. Little Friday night coding exercise to get this working.
Enjoy - and let me know of any missing functional groups! I could only do a few hundred.
Enjoy - and let me know of any missing functional groups! I could only do a few hundred.
Half of an AI scientist is rejecting or accepting hypotheses. FutureHouse and Science Machines just put out ~300 novel hypotheses from ~50 published papers along with ground-truth data. Humans take 4.2 hours to solve these and frontier models get 10-20% correct.
This is like SWE-bench for comp bio
This is like SWE-bench for comp bio
March 4, 2025 at 4:41 PM
Half of an AI scientist is rejecting or accepting hypotheses. FutureHouse and Science Machines just put out ~300 novel hypotheses from ~50 published papers along with ground-truth data. Humans take 4.2 hours to solve these and frontier models get 10-20% correct.
This is like SWE-bench for comp bio
This is like SWE-bench for comp bio
PaperQA2 can now work with clinical trials. It considers both research papers and clinical trials jointly to answer complex questions. It uses the the clinicial trials dot gov API - so it can do complex queries too. Checkout the tutorial below:
futurehouse.gitbook.io/futurehouse-...
futurehouse.gitbook.io/futurehouse-...
February 25, 2025 at 4:27 PM
PaperQA2 can now work with clinical trials. It considers both research papers and clinical trials jointly to answer complex questions. It uses the the clinicial trials dot gov API - so it can do complex queries too. Checkout the tutorial below:
futurehouse.gitbook.io/futurehouse-...
futurehouse.gitbook.io/futurehouse-...
It's been about a month since the first batch of reasoning models was released. There’s been about a dozen reproductions since then and some patterns are emerging. I’ve written up my own notes on training recipes, frameworks, rumors, and major open questions.
diffuse.one/p/d2-000
diffuse.one/p/d2-000
February 22, 2025 at 8:09 PM
It's been about a month since the first batch of reasoning models was released. There’s been about a dozen reproductions since then and some patterns are emerging. I’ve written up my own notes on training recipes, frameworks, rumors, and major open questions.
diffuse.one/p/d2-000
diffuse.one/p/d2-000
This is still an early topic and my work is very preliminary, but I think we may be able to start auditing scientific literature at scale. I’ve written up a lot of thoughts, background, and analysis in a blog post: diffuse.one/p/d1-008 4/4
February 19, 2025 at 1:26 PM
This is still an early topic and my work is very preliminary, but I think we may be able to start auditing scientific literature at scale. I’ve written up a lot of thoughts, background, and analysis in a blog post: diffuse.one/p/d1-008 4/4