Luke Hutchison (@LH) profile picture

Computer scientist, computational biologist/chemist/physicist/cosmologist; PhD, MIT/Harvard; former Google AI research scientist; TED Fellow. Creating @ZedLang

Planet Earth
Joined May 2009
Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)Photo by Luke Hutchison (@LH)
After a year of intensive and difficult work, I completed and submitted two physics papers that identify the origin of all force laws (other than the weak and strong force), including gravity and inertia. This also identifies the origin of the equivalence principle. A core finding: time dilation creates gravity, gravity does not create time dilation. The central identity identified in the first paper, F = −ħ∇ω (the force experienced by a particle is linearly proportional to the gradient in the frequency of the particle's wave function) unifies general relativity (∇ω), quantum mechanics (ħ) and classical mechanics (F), without requiring the quantization of gravity or invoking a graviton. In the second paper, I reformulated the Dirac equation (the relativistic version of the Shrödinger equation for the wave function of a particle) into quaternionic form, and was able to show that all force laws other than the strong and weak force can be directly obtained from the wave function, in combination with the force-frequency identity of the first paper. Submitted to Physics Review D. PDF links: (1) cdn.jsdelivr.net/gh/lukehutc… (2) cdn.jsdelivr.net/gh/lukehutc…
16
22
3
240
619,847
Luke Hutchison retweeted
Se acabó... Ya tenemos GANADOR al mejor vídeo con IA de 2026. Así se descubrió el aceite.
🤖 Made with AI
327
1,484
84
18,619
2,686,552
Luke Hutchison retweeted
i fine-tuned qwen3-vl 8b to watch a video and tell you how the shot was made. i'm calling it shot reader. camera height, lens, angle, how the camera moves. who's in frame, where they stand, where they walk. turns out that's hard for ai. even opus 5.5 and gemini 4, with a prompt written specifically to reverse-engineer shots, get the camera height and move both right less than half the time. open models out of the box: 5%. shot reader gets 69%. it's 2.4x faster than opus, ~30x cheaper, and small enough to run on your own gpu. why it matters: ai can already write blender. getting a shot right starts with reading it. see a shot → get the plan → rebuild it in 3d, reshoot it, remix it. open weights on hugging face, link below.
69
61
14
917
2,023,939
Luke Hutchison retweeted
who made this ??!!! 😭 bored.com
80
450
60
4,649
801,849
Luke Hutchison retweeted
RSI is a process that's been happening at least since ~600,000 years ago
rsi is a process that’s been happening at least since the renaissance
29
60
4
871
46,590
Luke Hutchison retweeted
Another classic Tony Hoare quote on language design: "One way is to make it so simple that there are obviously no deficiences and the other way is to make it so complicated that there are no obvious deficiencies."
5
5
1
73
2,485
You’re not being told the truth, here is what happened. 🚨
2,120
5,059
328
20,362
1,025,036
Luke Hutchison retweeted
using carbon with 4 hydrogen subagents. i call it... methane.
147
537
85
8,925
172,105
Luke Hutchison retweeted
Most maps cut the ocean apart to keep the land whole. The Spilhaus projection splits the continents instead, so all of Earth's seawater reads as one body. A year of surface currents, colored by sea temperature. Now also a print. #ocean #dataviz
134
1,639
144
10,154
307,618
Hilariously the grok.ai domain is being squatted right now... I can only imagine the asking price
2
1
3
523
Luke Hutchison retweeted
why is using AI more exhausting than when we used to write code manually
154
20
14
536
33,762
I'm continuing to see that Neon beats all other frontier models including Fable 5.2 on hard algorithmic problems -- and it's not close. However, Neon also hallucinates more. Two or three more generations of model and there will be only one game left in town: Gemini.
3
1
15
882
Luke Hutchison retweeted
3
4
33
1,081
This matches my experience
Gemini 4 Neon and Gemini 4 Carbon: first impressions. I've been using these models for a while now and have some things to say: Gemini 4 Neon 1. The most creative model I've ever seen. Because it doesn't have the brakes that Fable 5.2 and GPT-6.2 Astra have, it can come up with more interesting solutions to hard problems than they do. I'd been working on a difficult problem with Fable 5.2 for 2 days and was stuck on it. But Neon came up with an absolutely brilliant solution. By the way, Opus 5.5 couldn't handle the task at all 2. You can tell it's a relatively small model. It lacks intellectual horsepower in some places. Because of that, it may not dig into the details as deeply as Fable does 3. Its execution level is comparable to Opus 5.5. Honestly, I don't see any difference between them in coding. Neon might even be better in places. It writes cleaner, more beautiful code 4. In conversation, Neon is practically indistinguishable from a real person. It writes the most beautiful and clear text. I tested it in Russian and English, and its writing is very pleasant to read. I'd say the best of anything on the market 5. On top of the text quality, I'd add its optimism and positivity. It's like your best bro. But sometimes this trait gets in its way on technical tasks. It can judge a potential result too positively and overlook the details and weak spots of its own solution 6. Understands a codebase extremely well. Good bug hunting. And a complete absence of slop in the code 7. If you need to come up with a marketing plan, a project idea, or some features, nobody will do it better than Gemini 4 Neon 8. Excellent taste. At the very least on par with Opus 5.5. But I haven't done much comparison on UI tasks 9. It can sometimes get lost in long tasks. I haven't tested this much, but I have a feeling it may suffer from context rot. I need to work with it a bit longer and test more 10. Very fast. On xhigh it runs at the same speed as Opus 5.5 on medium This is definitely one of my favorite models now. I want to work with it more and more Gemini 4 Carbon 1. In benchmarks Neon beats Carbon, but that's clearly not the real picture. Carbon is without a doubt the bigger and smarter model 2. More thorough and thoughtful than Neon. Give it a big, complex task, and it will autonomously investigate everything and come back with a detailed, thorough, high-quality result. I can't say it's better than Fable 5.2, but it's definitely no worse. As a researcher and analyst specifically, they're on the same level 3. Carbon is 100% better than GPT-6.2 Astra. Literally at everything. The gap is like the one between Fable and Sonnet. And I'm not exaggerating 4. Very, very, very slow. I haven't seen a model this slow in a long time. Ask it the simplest question, something like "How are you," and it'll come back with an answer in about 5 minutes 5. Because of how slow it is, I wouldn't trust it with coding tasks, and there's no point anyway. Neon writes better code 6. Otherwise it's the same as Neon. All of Neon's strengths apply to Carbon too. I won't list them again If you need to plan a big, complex project, you can confidently turn to Carbon. Thanks to its creativity and good writing, in most cases it will give you a more interesting result than Fable 5.2, which is heavily stifled by its system prompt and distillation
2
4
1
290
80,826
Luke Hutchison retweeted
I have been programming for over 40 years, since I was a kid. AI has been able to generate software for the last 2 years, but it didn't do a great job. I swore I would never use AI to write any code that I cared about: I care a LOT about code quality, and AI just couldn't do as good a job as I could. That all changed 2 months ago with the release of Claude Opus 5.5. Opus can now create any sort of software better than 95% of human programmers, and 1000x faster. Does AI take away my job or yours? No, only if you believe it will, and you give up, or you get lazy by offloading work to AI and not doing any cognitive heavy lifting yourself. AI is a power tool for the human brain. AI gives humans superpowers. Now creativity, wisdom, and tenacity are the key skills for any sort of knowledge work, because those are the skills that you need to wield this new tool well. As of two months ago, anyone can create anything that they can dream of: any app, any website, any business, even any hardware product. All the barriers to entry are gone. You no longer have to spend 10-40 years of your life trying to get really good at engineering: we all now have an army of AI engineers at our command. Am I sad that the decades I invested in computer science and software engineering, including 13 years of university studies, have been rendered irrelevant? No, because they're not at all irrelevant: I have loved every bit of the journey, and the wisdom I have gained along the way is now a critical enabler for all my future work. I could not love the new world more. The central question now is: **Can you dream big enough?**
142
156
25
1,411
1,081,160
Since when did CPUs cost $20,000?
1
7
494
Luke Hutchison retweeted
Gemini 4 Neon High vs Opus 5.5 High vs Sol 6.2 High "Create a bluetooth electric toothbrush using tscircuit..." there is a lot of potential here but it made a lot of mistakes (not finishing routing, poor footprint choice). However, it does have the best looking toothbrush
16
10
6
447
60,150
Luke Hutchison retweeted
Replying to @theo
We are glad you like it! We have more of where this came from and will keep shipping at the same speed.
8
4
2
327
28,349
Luke Hutchison retweeted
We invented a machine that can do anything you ask for. Now all we need to do is figure out what the right thing to ask for is
207
281
75
4,696
94,598