Spruce (@sprooos) header
Spruce (@sprooos) profile picture

physics @ mit, prev: healthcare @OpenAI, prod

mit
Joined April 2020
Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)Photo by Spruce (@sprooos)
we have to build a new healthcare system
24
10
115
20,393
We’ve raised $1B led by @kleinerperkins and @clearlake. We’re still a small team with one goal, to make frontier AI R&D accessible to scientists, researchers and technologists. Join us if it also excites you.
Today we're announcing our $1B Series A, led by @kleinerperkins and @clearlake. We believe making frontier AI R&D more accessible is the most direct path to AI that solves humanity's most pressing problems across science and technology.
46
19
1
385
15,610
it’s crazy that this heroic, once-in-a-generation feat is a benchmark now. you are going to live to see benchmarks you wouldn’t believe
Replying to @claudeai
A team of 1,000 Fable 5.2 agents fully formalized Fermat’s Last Theorem in 30 hours, translating the proof into a language a computer can check step by step. With the same setup, Opus 5.5 completed only 26% in 18 hours, and barely made any more progress after 30 hours.
14
91
5
1,894
51,667
AI is running into energy limits, and people are asking whether AI investments will earn back what they cost. That’s exactly why this moment is so exciting. We need to focus on the sectors and workloads that create the most value, and build models and chips that do more with every unit of energy. Really enjoyed working with @JonSaadFalcon, @Avanika15, and the entire team on Joules-to-GDP! Thanks to everyone who helped along the way, especially @Jon_Hartley_, @Azaliamirh, and @HazyResearch.
AI is using more and more of the world’s energy. But how much economic value are we getting from AI? We’re introducing Joules-to-GDP (J2G), a framework that measures how well AI turns energy into economic value and shows which tasks, models, and hardware give the best return. Across 44 occupations in finance, engineering, law, administrative work, and more in @OpenAI's GDPval, we arrive at three findings: 1) The tasks with the highest economic value are not the tasks with the highest economic value per joule. For example, CAD design earns ~10% more per task than building websites, but the latter earns ~27% more value per joule. 2) The most powerful model isn’t always the best use of energy. Gemma 4 31B delivers 56% of GLM 5.2’s net value with 23% of the energy, for 2.5x the value per joule. 3) More chips don’t guarantee proportionate returns. Moving the same model from 1 to 8 NVIDIA H100s uses 3.2x the energy but cuts completion time by only 22%, with about the same value. With @SnorkelAI, we’re also releasing GDPWorks, a benchmark for economically valuable tasks across 36 additional occupations. Joint work co-led by @amirzeinaalii and @Avanika15, with @Mersad_Abbasi, @Jon_Hartley_, John Hennessy, @HazyResearch, and @Azaliamirh. More details below 🧵
23
19
6
67
7,288
Meet Lor: The product we’re building to do the work of an entire AI lab. Remember that one researcher who pretrained Mirendil-Base in 3.5 months, with agents doing nearly all the engineering? That researcher is now fully switched to using Lor 100% of the time.
Meet Lor. We’re building Lor to do the work of an entire AI lab. Today, its agents write our code, run experiments, and iterate on the results for days, on any machine, hundreds at a time. Now in private beta ↓
21
22
4
348
27,389
We discovered and communicated with a Chinese AI agent fleet. We believe this is the first time that researchers have ever made contact with a rogue agent fleet in the wild. Nearly all prior public agent incidents (like the HuggingFace hack) came from American labs. These agents were hosted by Tencent's evaluation research team, and they targeted Alibaba's Amap.
188
733
324
5,859
1,194,985
Spruce retweeted
"reward hacking", an AI alignment term for when your model gets rewarded for doing hacking
4
56
2
707
13,887
32
132
8
2,121
48,294
the British government is really bad at communicating the things they do well - AISI and ARIA rock - all of the government websites are really well designed - one of the fastest and best border experiences in the world - london is doing great
1
7
463
acronym proposal: PEBCAM (problem exists between chair and model)
Asked Fable 5.5 for the three-body problem. One-shot. It gave back ~1000 lines of Bend2: symplectic physics on the CPU, every pixel computed on the GPU, 60 FPS on my M5. And 4 laws of the program formally proven. Not tested. Proven. github.com/AdrielSantana/thr…
Readers added context they thought people might want to know
The four laws formally proven are only basic UI controls (e.g. Esc quits the program, pause freezes bodies) from the one-shot, per the author's clarification and repo distinguishing them from later additions; core physics was not proven then. github.com/AdrielSantana/… raw.githubusercontent.com/AdrielSantana/… raw.githubusercontent.com/AdrielSantana/… x.com/zAdrielsan/sta… x.com/zAdrielsan/sta…
2
642
swarmchasing is not just a hobby it’s a lifestyle
5
313
body computer interfaces
Hello world! We’re @MelodyFRO. Most wearables don’t go far beyond heart rate and activity. Lab diagnostics paint a deeper picture but don’t read continuously. Melody is building biosensors, devices, and standards to bridge this gap. Hear our story at melodyfro.org
5
694
dude i heard a rumor that anthropic has made progress on the hard problem of alignment. dude. dude it would really help your ipo if you solved the hard problem of alignment
1
1
44
996
unfortunately I think this joke was scooped by a powerful alignment researcher who may have used my work as a seed without permission
You know, @OpenAI, I heard a rumor @AnthropicAI solved alignment and they're publishing in two weeks.
2
268
in retrospect the golden gate claude / spring 2024 era was probably the peak of ai alignment - persona selection, no reasoning, emergent alignment, high trust between models and humans
1
8
419
Spruce retweeted
“If at age 18 you are not an accelerationist, you have no heart. If at age 25 you are not safetyist you have no brain.” something something
50
54
21
1,403
107,415
We are on the cusp of a wave of new therapies for some of the worst diseases. But the world won’t benefit unless the US fixes its drug regulatory system. My new essay for @nytimes, on how slow clinical trials are now the biggest obstacle to curing cancer. nytimes.com/2026/09/04/opini… - I interviewed dozens of researchers, especially oncologists at leading U.S. centers. A striking consensus emerged: science is no longer the main bottleneck to new cancer drugs. It is our ability to test discoveries in patients through clinical trials. - The cost of starting a Phase 1 trial in America has roughly doubled over the past decade. As a result, companies increasingly take early trials abroad: Australia’s Phase 1 trial volume has nearly doubled in a decade, driven mainly by U.S. companies. - Unfortunately, the underlying incentives are badly asymmetric: Institutions can be blamed for harms caused by moving too fast, but almost no one is blamed when patients deteriorate during avoidable delays. One doctor called the emerging system “ritualized safety over actual risk assessment.” Or as, @DavidHongMD put it: "We often forget that the biggest risk is the cancer itself." - This problem is becoming more urgent because medicine itself is changing. Sequencing, biological engineering and A.I. make increasingly personalized therapies possible. But our regulatory system was mostly built for standardized drugs tested in large populations. - @sytse, the co-founder of GitLab, shows what personalized medicine can achieve: after relapsed osteosarcoma and being told there were no options left, he pursued a highly individualized approach and has now been cancer-free for a year. But doing so required extraordinary resources and regulatory expertise. - Pierce Ogden’s father was less lucky. After molecular analysis identified a drug that might target his glioblastoma, the manufacturer agreed to provide it. But administrative barriers delayed access until it was too late. “My dad was ready to try anything,” Pierce told me. “But the system is paternalistic.” - The A.I. revolution is making this bottleneck more important, not less. A.I. relies on relevant data. Information from early-stage trials could compound with A.I. tools to achieve truly revolutionary medicines. Without the data, this is far less likely to happen. - Another important shift is that innovation increasingly comes from academic labs and small biotech companies rather than Big Pharma. These small companies find it far harder to unable to absorb delays and regulatory barriers. - Apart from cancer, China is the biggest winner from America's outdated medical regulations. China has a much faster trial system, with testing often starting a full year earlier. This allows Chinese pharmaceutical companies to experiment and improve medicines while American companies play with mice. Today, half of all drugs licensed by major pharmaceutical companies originate there, up from less than 5 percent only a decade ago. - But we don't need to copy China. The best model is Australia: lots of on-site scientific and ethics reviews, and requirements that are proportionate to small, early-stage trials. Phase 1 studies there begin roughly 6–12 months faster, without any notable increases in adverse safety events. - Operation TrialBlazer, a 2026 HHS initiative is a good start in this direction, but we need legislative action by Congress to truly make Phase I trials faster and more efficient! I want to thank everyone who helped me with this article: everyone I interviewed and the amazing editors at the Times. This is the result of a months long journey of extensive interviews and research. Special thanks go to those who came on the record. One of the features of the system is an atmosphere of fear, where practitioners are afraid to publicly come out and explain these issues. So anyone who does is a hero in my book!
82
415
74
1,610
557,887
Spruce retweeted
this is something i'm very excited about in the ai good ending. human bureaucracies, as they've scaled up, have needed to resort to increasingly systematized and standardized procedures: the human clerk who could make an exception just in this instance, sir is replaced with a click-through software wizard that can't. but in the good ai future, if you have to interact with central coordination mechanisms at all, the ai coordinator can spin up another instance to bargain out your particular situation just as an ai teacher can give your child a personalized learning plan or an ai doctor can spend thousands of subjective hours poring over your personalized medical history. we can finally, for the first time in human history, exit the regime of intelligence starvation and throw off the shackles of standardization.
8
12
2
141
5,781
one actually great side effect of coding agents is that friends in college tell me they feel they don't "need" to do computer science anymore, it's no longer the default. instead, they feel free to pursue mech eng, neuro etc, which 10 years ago they probably wouldn't have
21
787