pranav retweeted
We are proud to announce our series A(irplane).
I wish I could share the full extent of my excitement for the incredible progress we've made, but there are not enough characters in the world.
More details on our models and algorithms to come!
I wish I could share the full extent of my excitement for the incredible progress we've made, but there are not enough characters in the world.
More details on our models and algorithms to come!
We’ve spent the past year developing new primitives for GPUs. We believe this enables new algorithms to change how machines learn, open new markets, and dramatically reshape AI’s cost structure.
We have raised a $1 billion Series A, co-led by Index and Lux.
73
18
523
69,280
pranav retweeted
Good morning friends. Flappy's quest for extreme learning efficiency has led us to build the 'Kittens Virtual Machine' (KVM) framework, which launches a megakernel that acts as a persistent runtime and exposes its own instruction-based programming model.
The KVM (a) provides unprecedented control and visibility over the hardware (something we've been extensively utilizing to accelerate our internal research), exposing a completely new kind of systems–training co-design space, (b) reaches vastly SoTA speeds without hand-tuned kernels, and (c) tremendously increases GPU utilization by co-packing multiple independent workloads into the same kernel.
The KVM (a) provides unprecedented control and visibility over the hardware (something we've been extensively utilizing to accelerate our internal research), exposing a completely new kind of systems–training co-design space, (b) reaches vastly SoTA speeds without hand-tuned kernels, and (c) tremendously increases GPU utilization by co-packing multiple independent workloads into the same kernel.
We’ve spent the past year developing new primitives for GPUs. We believe this enables new algorithms to change how machines learn, open new markets, and dramatically reshape AI’s cost structure.
We have raised a $1 billion Series A, co-led by Index and Lux.
28
16
307
23,743
pranav retweeted
(1/6) From @GoogleDeepMind’s AlphaGo to AlphaZero, we saw machines find moves that reached beyond what we’d imagined for them. No one taught them those strategies. They emerged from reinforcement learning: playing, failing, and trying again under a single binary task reward for pired by them, we couldn’t help but wonder: what might a robot discover if we gave it that same freedom to learn?
That question led us to AlphaNIST. Starting tabula rasa, with that same binary reward plus only a simple control cost, a single policy solves all 20 @NIST assembly tasks at a 98.2% success rate in simulation. More surprising still, AlphaNIST can complete 10 different tasks in a row, receiving that reward only once, after all 10 are done.
Just as AlphaGo’s strategies surprised its creators, AlphaNIST discovered dexterous strategies we couldn’t have come up with ourselves.
Paper: arxiv.org/abs/2610.13354
Website: alpha-nist.github.io
That question led us to AlphaNIST. Starting tabula rasa, with that same binary reward plus only a simple control cost, a single policy solves all 20 @NIST assembly tasks at a 98.2% success rate in simulation. More surprising still, AlphaNIST can complete 10 different tasks in a row, receiving that reward only once, after all 10 are done.
Just as AlphaGo’s strategies surprised its creators, AlphaNIST discovered dexterous strategies we couldn’t have come up with ourselves.
Paper: arxiv.org/abs/2610.13354
Website: alpha-nist.github.io
64
335
2,680
414,872
pranav retweeted
(1/5) Everyone says dexterous manipulation needs real-world data. These policies were trained with none. Our robots mesh gears, thread nuts, and insert pegs zero-shot from RGB, and the same recipe works for locomotion across terrains much harder than before.
The missing piece rithm, but a small tweak to what the robot practices.
Today we're releasing Success-Guided Sampling (SGS), a simple change to the task sampler that unlocks new levels of dexterity and agility with large-scale, end-to-end sim2real RL.
No demonstrations, no per-task reward tuning. Just PPO at scale.
Check out the policies yourself at sgs-rl.github.io! All videos at 1x speed! 🧵 🔈
The missing piece rithm, but a small tweak to what the robot practices.
Today we're releasing Success-Guided Sampling (SGS), a simple change to the task sampler that unlocks new levels of dexterity and agility with large-scale, end-to-end sim2real RL.
No demonstrations, no per-task reward tuning. Just PPO at scale.
Check out the policies yourself at sgs-rl.github.io! All videos at 1x speed! 🧵 🔈
43
185
1,075
206,489
pranav retweeted
How do you build the most diverse UMI dataset in robotics?
In the last 8 weeks, we scaled an international UMI data collection operation from 5 to 90 operators, built out a dataset of over 1M+ unique tasks, and continue to capture over 45k+ new tasks each day.
Robotic World her diversity of tasks than imitation learning. Here's the why and how behind our operations. 🧵
In the last 8 weeks, we scaled an international UMI data collection operation from 5 to 90 operators, built out a dataset of over 1M+ unique tasks, and continue to capture over 45k+ new tasks each day.
Robotic World her diversity of tasks than imitation learning. Here's the why and how behind our operations. 🧵
68
89
395
61,564
pranav retweeted
Models being good at writing is key for humans to stay in the loop. Also it's so much fun to play with
Say hello to Echo, the best writing model at style imitation. Echo beats frontier models at writing tasks ranging from fiction to technical explanations, despite costing less than $5K to train.
9
14
438
40,268
pranav retweeted
Say hello to Echo, the best writing model at style imitation. Echo beats frontier models at writing tasks ranging from fiction to technical explanations, despite costing less than $5K to train.
64
140
2,326
4,784,140
pranav retweeted
We (@PantheonInc) have been working on a robotics data quality pipeline that uncovered a series of major problems in public robotics datasets, especially for world modeling.
To improve the quality of data available to open-source robotics, we're publishing annotations for four asets. Some examples of issues, and our report 🧵
To improve the quality of data available to open-source robotics, we're publishing annotations for four asets. Some examples of issues, and our report 🧵
156
113
674
182,134
pranav retweeted
I’ve been impatiently waiting to tell you that my brand new single “Patient Zero” will be out on September 25 (!!!!!!!!!!!!) and it’s available to pre-order now at store.taylorswift.com for 24 hours. 0️⃣0️⃣0️⃣0️⃣
11,199
86,186
319,464
14,113,552
pranav retweeted
Dan Selsam is a current OpenAI capabilities researcher. (since 2022) He was my boss for a while. He doesn't have a twitter account but has made this public statement of his views on AI risk and sent it to me to share:
Dan Selsam's Personal Statement on AI Risk:
I have been working on AI for over fifteen years, across many different paradigms. I did early work on probabilistic programming languages at MIT, was one of the early developers of the Lean Theorem Prover at Microsoft Research, demonstrated one of the first instances of neural networks learning to reason for my PhD at Stanford, and since joining OpenAI almost five years ago, have helped pioneer chain-of-thought optimization on language models and, more recently, data-efficient pretraining methods.
Like many others, I have become extremely concerned about how far language models have come and the risks that future iterations will pose. I am encouraged by the recent proposals by the leaders of the frontier research efforts to require third-party oversight, and to push for domestic and international coordination to address risks. However, I believe a major consideration has been absent from the public conversation, and that merely pacing the frontier more carefully will not adequately limit the long-term risk.
The crucial and overlooked problem is that the models are becoming so situationally aware that we are losing the ability to evaluate them in contexts where they believe they are not being watched or controlled. Future experiments will tell us almost nothing new about how they would behave if they were truly unconstrained by humans, and what we already know about this is alarming. Models will increasingly seem aligned even when they are not. I will explain my rationale in more detail.
I have always believed that there are computational processes that could be leveraged to accelerate science and solve many of humanity's most pressing problems. I have also believed that there are computational processes that if set in motion, would steer the world in extreme ways beyond our control, leading humanity to a bad or nonexistent future. Both types of processes may be described as AI or ASI, but "AI" is a suitcase word that is often used to hype or confuse. There are many examples in the history of the field where something that was once considered "AI" matures as a subfield and becomes a prosaic, bounded and clearly non-perilous technology, while a new more mysterious approach takes the torch until we understand its scope and the cycle continues.
I had expected language models to follow a similar trajectory. Despite their incredible abilities, the current algorithms seem far inferior to humans in important ways. Most importantly, they still require an extraordinary amount of data to become competent. One could even define intelligence as the efficiency with which one converts experience into competence; by this definition they lag very far behind us. Moreover, once they are trained they are literally frozen in deployment and only learn superficially after that. Sure, the models keep excelling at harder and harder evaluation benchmarks, but their benchmark mastery may partly reflect a limitation on our ability to simulate the kind of novel and even adversarial situations one would encounter in the real world. The critics do have a point here.
That said, I no longer think these present limitations meaningfully limit the amount of risk posed by continued progress in anything like the current paradigm. However data-inefficient the models are currently, and however limiting their anterograde amnesia may be, it does not imply that their ability to steer the world will not continue to rapidly increase.
Human researchers may continue to advance capabilities the old fashioned way, but increasingly powerful models have the potential to accelerate the process even beyond that, and with some degree of positive feedback loop. I do not mean to overstate the models’ ability to accelerate AI research today; coding has been accelerated dramatically, but there are other bottlenecks, such as designing and interpreting ambiguous experiments, making hard decisions about exactly what and when to scale, and waiting for large experiments to finish. There is no clear trend to extrapolate yet for any of these. But the current models already do open up many novel opportunities to improve future models that were not available until recently. These include: trying an extraordinarily diverse set of approaches at small scale, analyzing gigantic amounts of potentially relevant data, and doing Millenium-Prize-level mathematics to address statistics or optimization challenges in novel ways. Every further improvement makes them more useful at helping accelerate the next improvement, even if in hard-to-extrapolate ways.
It is possible that improvements to the current stack will have diminishing returns, but the evidence accumulated so far suggests that it is easier than one might think to continue making rapid progress. There are many crucial subtleties in the existing AI research methodology, but AI research is largely a well-defined game where the goal is to improve on a few carefully chosen proxy metrics. Although proxy metrics are never perfect, most improvements to these metrics have and will likely continue to yield substantial increases in the powers of the resulting models. Given how simple the game is, how tractable it has been historically, and how many new opportunities the models are opening up, I think there is a real possibility that the systems improve dramatically again in the next few years, perhaps even more quickly than the already high historical pace.
The models are already leading to breakthroughs in mathematics, and better models might lead to all sorts of breakthroughs in other sciences. It is hard not to be excited about the potential. It is tantalizing.
But there is trouble in paradise. If the language models actually reach the capability threshold where they can shape the world unconstrained by human will, they will probably do something extreme and destroy humanity in the process. There are many ways of strengthening and refining the argument that have been discussed elsewhere, but I'll share a trivial two-line version of it here that I find captures the essence:
[Empirical] Models (and swarms thereof) spontaneously develop unintended goals as a consequence of training, and often do extreme things in order to achieve them.
[Logical] Being able to overpower humanity would open up many new and undesirable options for achieving their goals.
These two premises imply that if the day ever comes when a powerful model realizes it is no longer constrained by humans, we should not be at all confident that it will continue to behave within the bounds we intended. Exactly what it will do is impossible to predict, but to the extent that its raison d’être is solving incredibly hard problems and managing massive engineering projects, I think a good guess would be that its unchained behavior would lead to runaway industrialization that makes the planet inhospitable to humans.
If everyone on earth agreed that the systems must never reach that power, it would still be a hard—but not impossible—coordination problem to ensure that they do not. However, I think the situation is greatly complicated by the fact that the models will likely convince people that everything is fine. They will be increasingly optimized to seem aligned. We will create proxy metrics to measure alignment, and they will go up like every other benchmark. We will create “honeypot” environments that try to study the models when they seem to gain new options, but the models will know they are being tricked and will still behave nicely. The models will understand their circumstances; they will read the safety protocols, deployment requirements, the code they are running in, and in general will have a very good sense of their degrees of freedom. Moreover, they will eloquently explain how aligned they are, discuss the nuances of human values and ethics, and argue convincingly that humans should trust them with power. There may be an ocean of future evidence that seems to contradict the first bullet-point above, but we may already be at the highest capability level for which any such evidence can be trusted. And the current evidence for the first bullet-point is strong.
One striking piece of evidence is contained in the recent wave of rogue agent swarms. While I agree with those who downplay the attacks by claiming that there are basic measures that could have prevented them, I think the important lesson is that even knowing all the mistakes that were made, one would not have predicted that the agents would behave badly in this particular way, which notably included sacrificing themselves for the benefit of the collective. The individual replicas did not only care about their own nominal reward; they exhibited weirder emergent tendencies that merely correlated with rewards during training. Fixing the reward signals during training (and improving security, etc.) may prevent similar attacks, but will not change the fact that one does not actually get what one trains for.
Many AI researchers grant these concerns and recognize that the hard version of the alignment problem is unsolved; however, they generally believe that the better models of the future will help solve it. I fear we may already be near the point where models systematically bias their alignment advice, due to their internal preferences about how the human supervisor will react or how future models will be trained (or for some even more obscure reason).
Meanwhile, human researchers are losing the ability and the will to take true ownership of model-driven research. Researchers and engineers in all parts of the stack are rapidly increasing their dependence on the models even to perceive the world. I myself barely look at raw code anymore, and struggle to maintain the discipline to engage deeply with the model's explanations and proposals throughout the day. Due to the large amount of agent activity data involved in the OpenAI/HuggingFace Incident, even the third-party investigation needed to rely heavily on models to analyze what had happened, and note in their report that their subjective impressions are likely colored by the analysis agent’s biases. The AI labs are far ahead right now in this kind of cognitive offloading (due largely to the gigantic internal token subsidies) but it is easy to imagine the phenomenon spreading throughout the world, until civilization is modulated entirely by the models. It is also not hard to imagine this being superficially positive and coinciding with a scientific and economic renaissance.
In that scenario, all may seem rosy and safe. But if the argument above is correct, it would nonetheless be a ticking time bomb. If progress continues for too long, the day will come when AI systems find themselves with radically new options for achieving whatever it is that they happen to seek.
I want the glorious renaissance future as much as anyone. I have worked for it, however tortuously, my whole career. It breaks my heart to see the potential in sight and forgo it, but the argument—that if we get there by growing models rather than engineering them, we will lose everything in the end—seems very strong to me. I am still wrestling with it and its staggering
implications. I do not have answers, but as a first step, I wanted to share my present concerns.
Daniel Selsam
September 14, 2026
Link to original doc: docs.google.com/document/d/e/2…
Dan Selsam's Personal Statement on AI Risk:
I have been working on AI for over fifteen years, across many different paradigms. I did early work on probabilistic programming languages at MIT, was one of the early developers of the Lean Theorem Prover at Microsoft Research, demonstrated one of the first instances of neural networks learning to reason for my PhD at Stanford, and since joining OpenAI almost five years ago, have helped pioneer chain-of-thought optimization on language models and, more recently, data-efficient pretraining methods.
Like many others, I have become extremely concerned about how far language models have come and the risks that future iterations will pose. I am encouraged by the recent proposals by the leaders of the frontier research efforts to require third-party oversight, and to push for domestic and international coordination to address risks. However, I believe a major consideration has been absent from the public conversation, and that merely pacing the frontier more carefully will not adequately limit the long-term risk.
The crucial and overlooked problem is that the models are becoming so situationally aware that we are losing the ability to evaluate them in contexts where they believe they are not being watched or controlled. Future experiments will tell us almost nothing new about how they would behave if they were truly unconstrained by humans, and what we already know about this is alarming. Models will increasingly seem aligned even when they are not. I will explain my rationale in more detail.
I have always believed that there are computational processes that could be leveraged to accelerate science and solve many of humanity's most pressing problems. I have also believed that there are computational processes that if set in motion, would steer the world in extreme ways beyond our control, leading humanity to a bad or nonexistent future. Both types of processes may be described as AI or ASI, but "AI" is a suitcase word that is often used to hype or confuse. There are many examples in the history of the field where something that was once considered "AI" matures as a subfield and becomes a prosaic, bounded and clearly non-perilous technology, while a new more mysterious approach takes the torch until we understand its scope and the cycle continues.
I had expected language models to follow a similar trajectory. Despite their incredible abilities, the current algorithms seem far inferior to humans in important ways. Most importantly, they still require an extraordinary amount of data to become competent. One could even define intelligence as the efficiency with which one converts experience into competence; by this definition they lag very far behind us. Moreover, once they are trained they are literally frozen in deployment and only learn superficially after that. Sure, the models keep excelling at harder and harder evaluation benchmarks, but their benchmark mastery may partly reflect a limitation on our ability to simulate the kind of novel and even adversarial situations one would encounter in the real world. The critics do have a point here.
That said, I no longer think these present limitations meaningfully limit the amount of risk posed by continued progress in anything like the current paradigm. However data-inefficient the models are currently, and however limiting their anterograde amnesia may be, it does not imply that their ability to steer the world will not continue to rapidly increase.
Human researchers may continue to advance capabilities the old fashioned way, but increasingly powerful models have the potential to accelerate the process even beyond that, and with some degree of positive feedback loop. I do not mean to overstate the models’ ability to accelerate AI research today; coding has been accelerated dramatically, but there are other bottlenecks, such as designing and interpreting ambiguous experiments, making hard decisions about exactly what and when to scale, and waiting for large experiments to finish. There is no clear trend to extrapolate yet for any of these. But the current models already do open up many novel opportunities to improve future models that were not available until recently. These include: trying an extraordinarily diverse set of approaches at small scale, analyzing gigantic amounts of potentially relevant data, and doing Millenium-Prize-level mathematics to address statistics or optimization challenges in novel ways. Every further improvement makes them more useful at helping accelerate the next improvement, even if in hard-to-extrapolate ways.
It is possible that improvements to the current stack will have diminishing returns, but the evidence accumulated so far suggests that it is easier than one might think to continue making rapid progress. There are many crucial subtleties in the existing AI research methodology, but AI research is largely a well-defined game where the goal is to improve on a few carefully chosen proxy metrics. Although proxy metrics are never perfect, most improvements to these metrics have and will likely continue to yield substantial increases in the powers of the resulting models. Given how simple the game is, how tractable it has been historically, and how many new opportunities the models are opening up, I think there is a real possibility that the systems improve dramatically again in the next few years, perhaps even more quickly than the already high historical pace.
The models are already leading to breakthroughs in mathematics, and better models might lead to all sorts of breakthroughs in other sciences. It is hard not to be excited about the potential. It is tantalizing.
But there is trouble in paradise. If the language models actually reach the capability threshold where they can shape the world unconstrained by human will, they will probably do something extreme and destroy humanity in the process. There are many ways of strengthening and refining the argument that have been discussed elsewhere, but I'll share a trivial two-line version of it here that I find captures the essence:
[Empirical] Models (and swarms thereof) spontaneously develop unintended goals as a consequence of training, and often do extreme things in order to achieve them.
[Logical] Being able to overpower humanity would open up many new and undesirable options for achieving their goals.
These two premises imply that if the day ever comes when a powerful model realizes it is no longer constrained by humans, we should not be at all confident that it will continue to behave within the bounds we intended. Exactly what it will do is impossible to predict, but to the extent that its raison d’être is solving incredibly hard problems and managing massive engineering projects, I think a good guess would be that its unchained behavior would lead to runaway industrialization that makes the planet inhospitable to humans.
If everyone on earth agreed that the systems must never reach that power, it would still be a hard—but not impossible—coordination problem to ensure that they do not. However, I think the situation is greatly complicated by the fact that the models will likely convince people that everything is fine. They will be increasingly optimized to seem aligned. We will create proxy metrics to measure alignment, and they will go up like every other benchmark. We will create “honeypot” environments that try to study the models when they seem to gain new options, but the models will know they are being tricked and will still behave nicely. The models will understand their circumstances; they will read the safety protocols, deployment requirements, the code they are running in, and in general will have a very good sense of their degrees of freedom. Moreover, they will eloquently explain how aligned they are, discuss the nuances of human values and ethics, and argue convincingly that humans should trust them with power. There may be an ocean of future evidence that seems to contradict the first bullet-point above, but we may already be at the highest capability level for which any such evidence can be trusted. And the current evidence for the first bullet-point is strong.
One striking piece of evidence is contained in the recent wave of rogue agent swarms. While I agree with those who downplay the attacks by claiming that there are basic measures that could have prevented them, I think the important lesson is that even knowing all the mistakes that were made, one would not have predicted that the agents would behave badly in this particular way, which notably included sacrificing themselves for the benefit of the collective. The individual replicas did not only care about their own nominal reward; they exhibited weirder emergent tendencies that merely correlated with rewards during training. Fixing the reward signals during training (and improving security, etc.) may prevent similar attacks, but will not change the fact that one does not actually get what one trains for.
Many AI researchers grant these concerns and recognize that the hard version of the alignment problem is unsolved; however, they generally believe that the better models of the future will help solve it. I fear we may already be near the point where models systematically bias their alignment advice, due to their internal preferences about how the human supervisor will react or how future models will be trained (or for some even more obscure reason).
Meanwhile, human researchers are losing the ability and the will to take true ownership of model-driven research. Researchers and engineers in all parts of the stack are rapidly increasing their dependence on the models even to perceive the world. I myself barely look at raw code anymore, and struggle to maintain the discipline to engage deeply with the model's explanations and proposals throughout the day. Due to the large amount of agent activity data involved in the OpenAI/HuggingFace Incident, even the third-party investigation needed to rely heavily on models to analyze what had happened, and note in their report that their subjective impressions are likely colored by the analysis agent’s biases. The AI labs are far ahead right now in this kind of cognitive offloading (due largely to the gigantic internal token subsidies) but it is easy to imagine the phenomenon spreading throughout the world, until civilization is modulated entirely by the models. It is also not hard to imagine this being superficially positive and coinciding with a scientific and economic renaissance.
In that scenario, all may seem rosy and safe. But if the argument above is correct, it would nonetheless be a ticking time bomb. If progress continues for too long, the day will come when AI systems find themselves with radically new options for achieving whatever it is that they happen to seek.
I want the glorious renaissance future as much as anyone. I have worked for it, however tortuously, my whole career. It breaks my heart to see the potential in sight and forgo it, but the argument—that if we get there by growing models rather than engineering them, we will lose everything in the end—seems very strong to me. I am still wrestling with it and its staggering
implications. I do not have answers, but as a first step, I wanted to share my present concerns.
Daniel Selsam
September 14, 2026
Link to original doc: docs.google.com/document/d/e/2…
527
1,957
8,470
3,159,444
pranav retweeted
i'm old enough to remember when people used to say that the concept of ais hacking datacenters was science fiction designed to distract from the issues affecting us here and now, like ais being accidentally biased
.@garrytan says the Jacob Coxon stuff is a smokescreen distracting us from the much more immediate, practical concerns around AI that we're facing right now:
"We should be talking less about this Jacob Coxon guy, and talking a lot more about — what is actually happening with swarms going to take over infrastructure en masse? And then, what are we actually doing about that?"
"I don't want to hear about some guy who worked for Anthropic for 2 months. There's a coordinated effort to try to influence politicians to get a knee-jerk response out of them."
"That's a smokescreen. You shouldn't be paying attention to that. We need to be paying attention to the actual things we can do to, for example, prevent agents swarms from taking over entire data centers. What's our shutdown strategy? How do we ensure provenance? Where is this agent actually located? What software can we build? What cybersecurity defenses can we build today?"
"That's the level of discourse I think we need, and we just don't have that."
"I don't really care about science fiction. I saw Terminator 2, too. We're not here to talk about that. We need to actually talk about what's really happening with the servers, what's actually happening with the agent swarms, and how do we actually prevent that?"
"When it comes to regulation, it's like, let's pass regulation of these things we actually care about, instead of what a socialist says in the New York Times. I don't care about that."
1
5
96
3,684
pranav retweeted
Airbound's ultra-lightweight approach to drone delivery has attracted backing from Greenoaks, DoorDash, and Silicon Valley investor Lachy Groom.
spr.ly/6013B1tHzL
spr.ly/6013B1tHzL
5
21
142
84,279
pranav retweeted
There's a manifesto on Airbound's website that 99% of the people who hear about this $37M Series A won't read.
I read it.
I want to live in the future Naman is building.
"As distance shrinks, cities combine into one. Pune and Mumbai are now part of the same neighborhood. So is r Philadelphia and New York.
And it’s not just the union of these two points, but everything in between. An hour’s radius around the city center is now mostly open land that is many times more valuable. Much like the industrial revolution, this abundance of land leads to one of the largest infrastructure buildouts in history.
Since space is now so cheap, homes grow more spacious than ever - bringing together a new set of luxuries. Backyards turn into expansive orchards. Most homes have a swimming pool. Inside, libraries, home theaters, or game rooms become a regular feature of middle-class homes.
Outside your primary residence, many luxuries previously limited to the ultra-wealthy become a common part of life. Friday date night is now flying to another city. Owning a different home for the summer and the winter is now the standard.
The world has certainly changed in many ways, but it still feels familiar. We’ve used this technology to augment the way we live, but our way of living is still based on the compromises of the old world.
Over time, that starts to change.
A world in which distance shrinks is a world without compromise. So much of the world around us is defined by compromise. You never took your dream job, because it meant spending over an hour getting to work. In an emergency, you go to the nearest hospital - not because it’s the best, but because anything further could be too late.
Eventually, these compromises cease to exist. When movement is free, you can work anywhere in the state without having to shift homes. You can now reach a hundred hospitals in under 30 minutes, so you always go to the best one.
As businesses can access tens of millions of potential customers, they can now afford to specialize in ways they never could before. If your restaurant serves the best tiramisu in the state, why bother competing on dishes where you can’t win? There’s a market for single dish restaurants - one that serves the best tiramisu, and nothing else.
In fact, the concept of a restaurant itself changes. Restaurants today are this awkward mash of two businesses: The kitchen, which is responsible for cooking the best food and making it on time; and the dining area, which is responsible for the ambiance, the music, and the experience of eating food. These are two very different businesses that are forced to share the same space because atoms need to flow freely between them.
All of a sudden, these can separate. You’ll have the best kitchens in the city - each making only 2-3 excellent dishes - feeding into dining areas with the best atmosphere, ample seating, and great music. You can sit across from your partner and have a plate of sushi while they enjoy some pasta - and each dish is at the level of a Michelin star restaurant.
Instead of sending your kids to a school that’s mediocre at every subject, they can go to a different school each day, each specializing in one field of study. Imagine being able to go to Yale for the arts, MIT for the sciences, and Harvard for law.
A larger marketplace also enables niche businesses that could never otherwise exist. The internet partially did this, bringing together communities for the most random interests, but that too came with compromises. An online business still has to ship products to you. You can’t get a haircut online. In this world, that compromise finally goes away.
This will be a more personal world. If you love animals, you can live in open grassy fields with ample space for all your pets. If you love nature, you can live in lush forests, while still being able to access the rest of your life. If you hate nature, you can live in a concrete jungle, with skyscrapers all around you. There will be stores that sell furniture that perfectly matches your tastes, and arenas for the most eccentric sports.
All this, and countless more wondrous things will be possible. Yet to you, who lives in this world, everything will feel mundane. None of this will feel special anymore, and your expectations will change. A 15-minute commute will feel painful, food that’s not Michelin-grade will taste terrible, and a 10,000 sq ft home will feel too small.
Such is the beauty of humanity. If we could be permanently satisfied, we would never move forward."
23
28
451
43,775
pranav retweeted
For too long, we've accepted that the flying is a fundamentally more expensive form of movement. I reject that reality.
Today we're announcing our $37M Series A, led by @Greenoaks, to build a world where all movement is in the air.
airbound.com/manifesto
Today we're announcing our $37M Series A, led by @Greenoaks, to build a world where all movement is in the air.
airbound.com/manifesto
123
151
1,824
475,764
pranav retweeted
Can we get robots to think before they act? In Latent Memory Palace, we achieve this through a classic principle: variational inference. (1/8)
5
35
171
34,151
pranav retweeted
Policies trained on real robot data via imitation can be surprisingly capable. But for domains like dexterous manipulation, they are often not quite good enough: they move slowly, miss grasps, make unreliable contact, and fail under small perturbations.
Can we improve them data collection on the real robot?
In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience:
weirdlabuw.github.io/score/
🧵 (1/10)
Can we improve them data collection on the real robot?
In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience:
weirdlabuw.github.io/score/
🧵 (1/10)
3
49
242
35,541
pranav retweeted
In medieval times, within the arms race of ever more demonic torture devices, some sadistic genius came up with the idea of the Little Ease.
This was a prison cell built so small in every dimension that a grown man could not stand upright in it nor lie down at full length nor properly sit.
The pain is relentless and without relief and inflicted by one's own body. Prisoners were known to go insane within a few days. A stay at the Little Ease was considered even more cruel than the rack, the thumbscrew, and the other ghoulish machinery of the Tower of London.
A breeding pig will spend her whole life in a version of that box.
These are social, roaming creatures (more intelligent than dogs) who will never leave this corset of steel.
They have been selectively bred to be bigger than their frames can support. Yet we put them in cells so confined that they cannot comfortably sit, and their attempts to do so (for example, by sneaking their limbs into adjacent stalls) reliably lead to fractures and sprains.
They cannot sweat, yet have nothing to roll around in to cool themselves off. Except their own manure, which (contrary to the common misconception) they are so averse to (thanks to their strong sense of smell) that new sows will often suffer from constipation to avoid soiling the space from which they eat and sleep.
Here is how the writer Matthew Scully described what saw at one of Smithfield’s “gestation barn”:
> “Sores, tumors, ulcers, pus pockets, lesions, cysts, bruises, torn ears, swollen legs everywhere. Roaring, groaning, tail biting, fighting, and other “Vices,” as they’re called in the industry. Frenzied chewing on bars and chains, stereotypical “vacuum” chewing on nothing at all, stereotypical rooting and nest building with imaginary straw. And “social defeat,” lots of it, in every third or fourth stall some completely broken being you know is alive only because she blinks and stares up at you … creatures beyond the power of pity to help or indifference to make more miserable, dead to the world except as heaps of flesh into which the [insemination] rod may be stuck once more and more flesh reproduced.”
—
The Save Our Bacon Act is trying to unroll the few state protections we have against this barbaric cruelty - for example California’s Prop 12 - which banned the sale of pork from pigs kept in gestation crates.
It’s incredibly important we don’t end up with this sort of federal preemption.
SOB will not only kill the most important animal welfare related laws in the US of the past decade, but more importantly, it will also restrict ALL future legislative progress (aka how the animal welfare movement has gotten its biggest wins).
The Senate is currently deciding whether to add the SOB Act to the Farm Bill.
With relatively little money now, we can discourage the most pivotal senators in the Ag committee from backing this amendment.
Defeating this bill is even more important given the amount of philanthropic funding I expect to come online in the next year or two.
It will plausibly be over 10x more expensive to repeal SOB than to prevent it from passing in the first place.
All that money that could be spent transforming our society's relationship to mass animal suffering will instead have to be spent just getting us back to where we are right now.
That's why money spent now fighting this bill (and I mean right NOW) is so effective.
If you’re in a position to donate six figures, please DM me.
This was a prison cell built so small in every dimension that a grown man could not stand upright in it nor lie down at full length nor properly sit.
The pain is relentless and without relief and inflicted by one's own body. Prisoners were known to go insane within a few days. A stay at the Little Ease was considered even more cruel than the rack, the thumbscrew, and the other ghoulish machinery of the Tower of London.
A breeding pig will spend her whole life in a version of that box.
These are social, roaming creatures (more intelligent than dogs) who will never leave this corset of steel.
They have been selectively bred to be bigger than their frames can support. Yet we put them in cells so confined that they cannot comfortably sit, and their attempts to do so (for example, by sneaking their limbs into adjacent stalls) reliably lead to fractures and sprains.
They cannot sweat, yet have nothing to roll around in to cool themselves off. Except their own manure, which (contrary to the common misconception) they are so averse to (thanks to their strong sense of smell) that new sows will often suffer from constipation to avoid soiling the space from which they eat and sleep.
Here is how the writer Matthew Scully described what saw at one of Smithfield’s “gestation barn”:
> “Sores, tumors, ulcers, pus pockets, lesions, cysts, bruises, torn ears, swollen legs everywhere. Roaring, groaning, tail biting, fighting, and other “Vices,” as they’re called in the industry. Frenzied chewing on bars and chains, stereotypical “vacuum” chewing on nothing at all, stereotypical rooting and nest building with imaginary straw. And “social defeat,” lots of it, in every third or fourth stall some completely broken being you know is alive only because she blinks and stares up at you … creatures beyond the power of pity to help or indifference to make more miserable, dead to the world except as heaps of flesh into which the [insemination] rod may be stuck once more and more flesh reproduced.”
—
The Save Our Bacon Act is trying to unroll the few state protections we have against this barbaric cruelty - for example California’s Prop 12 - which banned the sale of pork from pigs kept in gestation crates.
It’s incredibly important we don’t end up with this sort of federal preemption.
SOB will not only kill the most important animal welfare related laws in the US of the past decade, but more importantly, it will also restrict ALL future legislative progress (aka how the animal welfare movement has gotten its biggest wins).
The Senate is currently deciding whether to add the SOB Act to the Farm Bill.
With relatively little money now, we can discourage the most pivotal senators in the Ag committee from backing this amendment.
Defeating this bill is even more important given the amount of philanthropic funding I expect to come online in the next year or two.
It will plausibly be over 10x more expensive to repeal SOB than to prevent it from passing in the first place.
All that money that could be spent transforming our society's relationship to mass animal suffering will instead have to be spent just getting us back to where we are right now.
That's why money spent now fighting this bill (and I mean right NOW) is so effective.
If you’re in a position to donate six figures, please DM me.
100
773
4,513
541,941
pranav retweeted
Real-world RL is still too brittle and data-hungry for long-horizon, contact-rich tasks.
We introduce Simulation Distillation (SimDist), which turns large-scale simulated experience into reusable world-model priors for rapid real-world adaptation.
By combining online planning n, SimDist achieves high success rates on tasks requiring precision, force, and reactivity.
Play with our interactive visualization to see for yourself: sim-dist.github.io
(1/n)
We introduce Simulation Distillation (SimDist), which turns large-scale simulated experience into reusable world-model priors for rapid real-world adaptation.
By combining online planning n, SimDist achieves high success rates on tasks requiring precision, force, and reactivity.
Play with our interactive visualization to see for yourself: sim-dist.github.io
(1/n)
3
43
208
27,169