Dileep George retweeted
Someone smart once told me that even when it feels like you have few options and little agency, you can always decide how you show up to a situation…model the behavior you wish you’d see more of.
I always find that to be an extraordinarily comforting and empowering thought
I always find that to be an extraordinarily comforting and empowering thought
3
6
57
2,488
Got nothing to do with America
After my bachelors I worked in Bangalore. First job. First time living in a city. (IIT Bombay campus doesn’t count)
A bunch of us from office went to a pizza place. My first time in a pizza place.
I pointed at something and I said I’ll have that pizza.
The look and correction I got from one of them is unforgettable “THAT’s LASAANIA”, which I later came to know is spelled lasagna.
After my bachelors I worked in Bangalore. First job. First time living in a city. (IIT Bombay campus doesn’t count)
A bunch of us from office went to a pizza place. My first time in a pizza place.
I pointed at something and I said I’ll have that pizza.
The look and correction I got from one of them is unforgettable “THAT’s LASAANIA”, which I later came to know is spelled lasagna.
i’m gonna make americans watch this scene to show how intimidating they can get
6
0
105
19,178
Thanks Prachee! Glad you liked #AGIComics! It now has a vibe-coded website too! agicomics.net
agree, the dirigibles blog from 2023 has remained surprisingly fresh: blog.dileeplearning.com/p/welcome-to-t…
agree, the dirigibles blog from 2023 has remained surprisingly fresh: blog.dileeplearning.com/p/welcome-to-t…
Helpful analogy and perspective by @dileeplearning who pulled off the impressive feat of writing about the state and future of AI in 2023 in a way that still feels relevant today!
Also TIL he also writes this delightful webcomic about AGI dileeplearning.github.io/comics_landing…
1
2
11
2,510
I've argued that the current era of AI is like the exciting era of dirigibles in 1920s -- flying by floating, without using the principles of aerodynamic lift. Conveniently, the timeline matches up exactly by 100 years, so maybe we can draw some parallels about mishaps too? At
6
0
43
3,447
If AGI came via the learn-by-experience route and LLMs didn’t exist, the first thing that AGI would do is invent LLMs.
10
2
38
3,473
Dileep George retweeted
Proud to share new work from our Paradigms of Intelligence team toward AI designed for decentralized hardware, where low-power communication and fault tolerance are critical requirements. Using Neural Cellular Automata (NCA), @mayalen_etc and her co-authors show that spatial locality is not a barrier to complex multi-step reasoning.
Read the full paper on arXiv: arxiv.org/abs/2609.36126…
Read the full paper on arXiv: arxiv.org/abs/2609.36126…
Can complex multi-step reasoning emerge purely from cells that only talk to immediate neighbors?
Happy to share our paper “Reasoning with Neural Cellular Automata (NCA)”, from our team at Google, Paradigms of Intelligence 🧵👇
6
22
147
6,049
The fact that Geoff didn't name it Schmidhuber Machine is a clue that he secretly admires @SchmidhuberAI 😇
If Ludwig Boltzmann was given a choice he’d have preferred the transformer architecture to be named after him rather than the Boltzmann machine.
0
2
12
1,615
If Ludwig Boltzmann was given a choice he’d have preferred the transformer architecture to be named after him rather than the Boltzmann machine.
5
1
23
3,601
Hmm. I did read the peer reviews and I’d call it peer review extra super lite. I’d be happy to do the job elife reviewers didn’t — I feel it shouldn’t be this joyous, stay tuned. :)
The @eLife review process was a joy as ever :)
3
0
32
7,795
I have one too! My name is on the same line as THE Nitesh's Tiwari's name in this honor roll! Sometimes life just lines you up with superstars, giving you eternal bragging rights. @niteshtiwari22
Aura farming time: THE Nitesh Tiwari has photographed me with two Filmfare award winners. His wife Ashwiny and his Lord Rama, Ranbir Kapoor.
6
0
51
26,399
Dileep George retweeted
FYI, for those wondering: we are working on organizing everything, and will be releasing data and code for our paper on Monday 10/26/26.
3
11
108
7,983
Dileep George retweeted
Check out our new model for creating realistic motion in a virtual rat during RL!
I think of this as providing the "brainstem and spinal cord" for NeuroAI researchers who want to simulate realistic neuroscience experiments.
Congrats to @aidan_sirbu for getting this out!
I think of this as providing the "brainstem and spinal cord" for NeuroAI researchers who want to simulate realistic neuroscience experiments.
Congrats to @aidan_sirbu for getting this out!
1/14) Virtual animals are a long-standing goal in neuro-AI, but they should actually behave like real animals.
We present SCAMPER: a virtual rat for RL studies that moves naturally during both exploration, and goal-directed behavior.
Project page: aidansirbu.com/SCAMPER-site/
3
9
70
6,239
Dileep George retweeted
We added Gemini 4 Argon to our AutoResearchExam Pareto curve. It is 2-3 times cheaper compared to the other frontier models for almost the same ML engineering performance. Congratulations to the GDM team.
14
28
349
43,874
congratulations on the launch @tarinziyaee . Super interesting approach. Good luck!
🚨 Personal news!
2000 years ago, Rome figured out a key idea Silicon Valley still hasn't.
Arenas can train - and monetize - the best embodied intelligences.
We're building embodied agents the same way.
Introducing Colosseums AI.
0
0
4
1,404
Dileep George retweeted
The ideal amount of healthcare is zero.
Progress ep09 is live with @annewoj23, co-founder of 23andMe.
We explore why "it's much cheaper if people die of a heart attack," running your own health with AI, and winning 23andMe back out of bankruptcy.
Progress ep09 is live with @annewoj23, co-founder of 23andMe.
We explore why "it's much cheaper if people die of a heart attack," running your own health with AI, and winning 23andMe back out of bankruptcy.
2
6
12
8,089
excited about what this team is up to!
@EkaRobotics has raised a $150M Series A.
Generality v/s Performance is the fundamental tension at the frontier of robotics. Generality without performance is just a prototype. Performance without generality is just a single-purpose tool.
Customers shouldn’t have to choose. Eka wered by our Vision-Force-Action (VFA) model, we deliver the performance and versatility customers need to put robots to work every day.
Eka is built on three convictions:
⚡️Force is the language of manipulation.
🚀 Reinforcement learning is how dexterous manipulation scales.
🎂 System 1—fast, reliable physical execution—is the bottleneck to solve.
With this funding, we will scale our model and accelerate customer deployments.
Grateful to be building alongside a dream team and investors who share our ambition.
3
1
20
2,193
Dileep George retweeted
@EkaRobotics has raised a $150M Series A.
Generality v/s Performance is the fundamental tension at the frontier of robotics. Generality without performance is just a prototype. Performance without generality is just a single-purpose tool.
Customers shouldn’t have to choose. Eka wered by our Vision-Force-Action (VFA) model, we deliver the performance and versatility customers need to put robots to work every day.
Eka is built on three convictions:
⚡️Force is the language of manipulation.
🚀 Reinforcement learning is how dexterous manipulation scales.
🎂 System 1—fast, reliable physical execution—is the bottleneck to solve.
With this funding, we will scale our model and accelerate customer deployments.
Grateful to be building alongside a dream team and investors who share our ambition.
Generality v/s Performance is the fundamental tension at the frontier of robotics. Generality without performance is just a prototype. Performance without generality is just a single-purpose tool.
Customers shouldn’t have to choose. Eka wered by our Vision-Force-Action (VFA) model, we deliver the performance and versatility customers need to put robots to work every day.
Eka is built on three convictions:
⚡️Force is the language of manipulation.
🚀 Reinforcement learning is how dexterous manipulation scales.
🎂 System 1—fast, reliable physical execution—is the bottleneck to solve.
With this funding, we will scale our model and accelerate customer deployments.
Grateful to be building alongside a dream team and investors who share our ambition.
82
108
1,014
125,693