I do not write posts on X. I tweet links to posts on other platforms. I like and retweet (occasionally) I comment on friends' tweets (rarely) Follow me on...⬇️⬇️⬇️

My opinion of @elonmusk I like his cars (I own a 2015 S, and 2023 S), his rockets, his solar energy systems, and his satellite communication system. I also like his positions on open source and patents. But I very much disagree with him on a number of issues. I disagree with how he treats his scientists. Technology/product development may not need openness and publications to advance, but forward-looking *research* sure does, whether it's in AI, neural interfaces, material science, or whatever. Secrecy hampers progress and discourages talents from joining the effort. I also disagree with the hype. I mean, expressing an ambitious vision for the future is great. But telling the public blatantly false prediction ("AGI next year", "1 million robotaxis by 2020", "AGI will kill us all, let'spause",...) is very counterproductive (also illegal in some cases). More importantly, I think his public positions on many political issues, journalism, the media and the press, and academia, are not just wrong but dangerous for democracy, civilization, and human wellfare. Say what you want about "traditional media" but you can't really have reliable information without professional journalists working for a free and diverse press. Democracy can't exist without it, which is why only authoritarian ennemies of democracy rail against the media. Finally he doesn't seem to hesitate to disseminate batshit-crazy conspiracy theories as long as they serve his interests (e.g. boosting "PizzaGate", "illegal immigrants corrupt elections in the US", "person X is a pedo",...). One would expect a technological visionary to be a rationalist. Rationalism doesn't work without Truth. This has become particularly concerning since he bought himself a platform to disseminate his dangerous political opinions, conspiracy theories, and hype. He has been quite naïve about the difficulties of running a social network and the (legal) necessity of doing content moderation. One can claim to be a 1st Amendment absolutist, but a lot of content *must* be taken down by law, e.g. terrorist propaganda, child exploitation, blatant hate speech (in the EU and other regions). Then, there is dangerous disinformation that puts public health in danger or corrupts the democratic process. You have to moderate that too. Content moderation is a complicated problem whose best answer is not an attitude of total laissez-faire but a complex trade-off.

Over 80 technical papers published since January 2022. What about you? scholar.google.com/citations…

You know @elonmusk, I really feel sorry for you and your family to have been infected by a mind virus. But the best defense against a mind virus is intelligence, knowledge, and science. Fighting a mind virus by promoting other mental pathogens, like Trumpism, conspiracy theories, and attacks on science and the media, is not a good idea. This is like trying to cure diabetes by spreading ebola: it will certainly stop your diabetes, but not without killing you. The cure against mind viruses is the expansion knowledge through science, and the dissemination of reliable information. Disinformation causes mind viruses to spread. Distrust in decent media outlets causes mind viruses to spread. Distrust in science and scientists cause mind viruses to spread. The political movements you are currently supporting are squarely on the wrong side of history here.

Must have been my tweet in defense of Anthony Fauci. Elon's call for Fauci to be prosecuted and imprisoned is pretty high up on the scale of anti-science assholery. Apparently, mistreating scientists is habit, whether they work for him or not.

This is huge: Llama-v2 is open source, with a license that authorizes commercial use! This is going to change the landscape of the LLM market. Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers Pretrained and fine-tuned models are available with 7B, 13B and 70B parameters. Llama-2 website: ai.meta.com/llama/ Llama-2 paper: ai.meta.com/research/publica… A number of personalities from industry and academia have endorsed our open source approach: about.fb.com/news/2023/07/ll…

Your smartest friends aren't very smart.

X is a $44 billion propaganda machine. Yet it attempts to disguise itself as a defender of unfettered free speech, a source of factual information, and a substitute for professional journalism.

Join xAI if you can stand a boss who: - claims that what you are working on will be solved next year (no pressure). - claims that what you are working on will kill everyone and must be stopped or paused (yay, vacation for 6 months!). - claims to want a "maximally rigorous pursuit of the truth" but spews crazy-ass conspiracy theories on his own social platform.

Elon has apparently one genuine technical paper indexed by Google Scholar: "An Integrated Brain-Machine Interface Platform With Thousands of Channels" Published in the Journal of Medical Internet Research in 2019. The authors? "Elon Musk and Neuralink" I'm sure the scientists who hide behind this collective name are super happy about that. I just hope they won't die bitter and forgotten. jmir.org/2019/10/e16194

I like his cars, his rockets, his solar panels, and his satellite network. I very much dislike his vengeful politics, his conspiracy theories, and his hype.

To people who think "China is surpassing the US in AI" the correct thought is "Open source models are surpassing closed ones" See ⬇️⬇️⬇️

This dude has absolutely no idea WTF he's talking about. The total net worth of my former PhD students alone would make any silvery Wall Street dude turn green with envy.

The more likely scenario is that the US becomes more like Norway. High income, universal healthcare, free higher education, low poverty rate, lots of oil *and* lots of Teslas. That would be good for you, no?

People who devote their life and career to the study of science, society, and human history care about reality, about truth, about people, don't care much about money, and don't believe in dogma and conspiracy theories. What a surprise!

Funny how the American Far Right feels the need to claim that every election they lose is rigged, even outside the US, apparently. The French Far Right was not "denied a victory." It just did not win because most voters did not vote for them, thankfully.

- Engineer: I invented this new thing. I call it a ballpen 🖊️ - TwitterSphere: OMG, people could write horrible things with it, like misinformation, propaganda, hate speech. Ban it now! - Writing Doomers: imagine if everyone can get a ballpen. This could destroy society. There should be a law against using ballpen to write hate speech. regulate ballpens now! - Pencil industry mogul: yeah, ballpens are very dangerous. Unlike pencil writing which is erasable, ballpen writing stays forever. Government should require a license for pen manufacturers.

Nice job! Open research / open source accelerates progress.

😅 you know, the guy with a good Twitter game 😅

So much misunderstanding of this comment! Here is a list of things I am *NOT* saying: - you need a PhD to do Science. You don't. A PhD teaches you to do research, but you can learn that on your own (though it's much easier with a mentor). - you need to get papers accepted by a journal or conference to publish: you don't. You can just post it in ArXiv.org . Many influential papers never went through the formal peer review process, or went through it after they became influential. - engineering is not science: it can be, depending on your methodology. I'm a scientist *and* an engineer. These activities are complementary and need each other. - science requires formal papers: it doesn't. A clear explanation on a website and a piece of code on a public repo will do. What I *AM* saying is that science progresses through the collision of ideas, verification, analysis, reproduction, and improvements. If you don't publish your research *in some way* your research will likely have no impact.

To qualify as Science a piece of research must be correct and reproducible. To be correct and reproducible, it must be described in sufficient details in a publication. To be 'published' (to receive a seal of approval) the publication must be checked for correctness by reviewers. To be reproduced, the publication must be widely available to the community and sufficiently interesting. If you do research and don't publish, it's not Science. Without peer review and reproducibility, chances are your methodology was flawed and you fooled yourself into thinking you did something great. No one will ever hear about your work. No one will pick it up and build on top of it. No one will build new technology and products with it. Your work will have been in vain. You'll die bitter and forgotten. If you never published your research but somehow developed it into a product, you might die rich. But you'll still be a bit bitter and largely forgotten.

* Language is low bandwidth: less than 12 bytes/second. A person can read 270 words/minutes, or 4.5 words/second, which is 12 bytes/s (assuming 2 bytes per token and 0.75 words per token). A modern LLM is typically trained with 1x10^13 two-byte tokens, which is 2x10^13 bytes. This would take about 100,000 years for a person to read (at 12 hours a day). * Vision is much higher bandwidth: about 20MB/s. Each of the two optical nerves has 1 million nerve fibers, each carrying about 10 bytes per second. A 4 year-old child has been awake a total 16,000 hours, which translates into 1x10^15 bytes. In other words: - The data bandwidth of visual perception is roughly 16 million times higher than the data bandwidth of written (or spoken) language. - In a mere 4 years, a child has seen 50 times more data than the biggest LLMs trained on all the text publicly available on the internet. This tells us three things: 1. Yes, text is redundant, and visual signals in the optical nerves are even more redundant (despite being 100x compressed versions of the photoreceptor outputs in the retina). But redundancy in data is *precisely* what we need for Self-Supervised Learning to capture the structure of the data. The more redundancy, the better for SSL. 2. Most of human knowledge (and almost all of animal knowledge) comes from our sensory experience of the physical world. Language is the icing on the cake. We need the cake to support the icing. 3. There is *absolutely no way in hell* we will ever reach human-level AI without getting machines to learn from high-bandwidth sensory inputs, such as vision. Yes, humans can get smart without vision, even pretty smart without vision and audition. But not without touch. Touch is pretty high bandwidth, too.

A reminder that people can disagree about important things but still be good friends.

You're being played by people who want regulatory capture. They are scaring everyone with dubious studies so that open source models are regulated out of existence.

To those who say 'these are *just* papers': One of these papers introduced convolutional neural networks (ConvNets) in 1989. Every single driving assistance system today uses ConvNets. That includes MobilEye (since 2014), Nvidia, Tesla, and just about everyone else. Technological marvels don't just pop out of the vacuum. They are built on years (sometimes decades) of scientific research that makes them possible. Research ideas and results are shared through technical papers. Without this sharing of scientific information, technological progress would slow to a crawl. scholar.google.com/citations… piped.video/FwFduRA_L6Q?si=sMox…

Now you're acting as if you were my boss 😂😂😂

WhatsApp vs iMessage is the new metric vs imperial. The entire world uses WhatsApp, save a few people who are iPhone-clutching Americans or from a country where WhatsApp is banned (like China). Yet iMessage users will actually claim it's objectively better, just like they will claim that Farenheit is intrinsically better than Celsius and inch better than cm, whereas it's just that they grew up with it.

Fun memories.

Very cool! Congrats.

Current LLMs are trained on text data that would take 20,000 years for a human to read. And still, they haven't learned that if A is the same as B, then B is the same as A. Humans get a lot smarter than that with comparatively little training data. Even corvids, parrots, dogs, and octopuses get smarter than that very, very quickly, with only 2 billion neurons and a few trillion "parameters."

I'm old enough to remember that in France in the 1960s and 1970s, far-right chauvinists were accusing Portugese immigrants of eating cats. It's an old trope. Today, making such an accusation publicly against an ethnic group violates laws against "incitement to racial hatred" in most of Europe. You may get sued, get a fine, and get blocked from running for office. Elon and other American 1st Amendment maximalists will scream "tyranny!" But anti-hate speech laws are the way European countries have learned to protect themselves against fascist takeovers. And they sure know what they're talking about.

X is not a source. At best, it's a relay. In practice, it's largely a propaganda megaphone for people like you, relayed by your fanbois and fanbots. In any case, it reflects the increasingly one-sided bias of its users. And yes, the press (finally) covers Trump negatively, like it should cover every felonious, lying egomaniac who just ridiculed himself in a debate on national TV.

The mark of the authoritarian: - Undermine the trust of the public in the sources of reliable information: the media, professional journalists, Wikipedia, universities, scientists, and government agencies. - Accuse those sources of being corrupt, woke, Marxist, fascist, beholden to powerful interests, stupid, pedophile, whatever. - Once the public is thoroughly brainwashed, fill their minds with ridiculous mind viruses: the country is bankrupt, it's being overrun by barbaric, murderous, subhumans immigrants who eat your pets and rape your daughters, our crypto-communist opponents want to take away your freedom of speech, elections are rigged, etc - Now claim: "only I can save the country from the coming Armageddon." This is a completely standard chapter from the authoritarian playbook. Almost half of Americans are falling for it, with the help of a few billionaires who are shamelessly willing to sacrifice democracy in the altar of their continued financial and political power.

Objectively, *you* are the biggest threat to democracy in America today. You are an oligarch who bought himself a huge megaphone so he could use it to support a would-be dictator, and to discredit the press, scientists, research organizations, the electoral process, and all the institutions necessary for a functioning democracy.

If you are a student interested in building the next generation of AI systems, don't work on LLMs

It seems to me that before "urgently figuring out how to control AI systems much smarter than us" we need to have the beginning of a hint of a design for a system smarter than a house cat. Such a sense of urgency reveals an extremely distorted view of reality. No wonder the more based members of the organization seeked to marginalize the superalignment group. It's as if someone had said in 1925 "we urgently need to figure out how to control aircrafts that can transport hundreds of passengers at near the speed of the sound over the oceans." It would have been difficult to make long-haul passenger jets safe before the turbojet was invented and before any aircraft had crossed the atlantic non-stop. Yet, we can now fly halfway around the world on twin-engine jets in complete safety. It didn't require some sort of magical recipe for safety. It took decades of careful engineering and iterative refinements. The process will be similar for intelligent systems. It will take years for them to get as smart as cats, and more years to get as smart as humans, let alone smarter (don't confuse the superhuman knowledge accumulation and retrieval abilities of current LLMs with actual intelligence). It will take years for them to be deployed and fine-tuned for efficiency and safety as they are made smarter and smarter.

Congrats John and Geoff! Both are former colleagues. I did my postdoc in Geoff's lab in Toronto. After that, I joined Bell Labs, where John was a part-time scientist (and a professor at Caltech). In fact, John was the reason why the department I joined was working on neural nets. I first met both of them in 1985. I met John at a workshop in Les Houches in early 1985. I met Geoff at the Cognitiva conference in Paris in June 1985.

🥁 Llama3 is out 🥁 8B and 70B models available today. 8k context length. Trained with 15 trillion tokens on a custom-built 24k GPU cluster. Great performance on various benchmarks, with Llam3-8B doing better than Llama2-70B in some cases. More versions are coming over the next few months. llama.meta.com/llama3/

The kind of attacks that Dr Anthony Fauci is being subjected to by Republicans is disgraceful and dangerous. This man is a hero. He helped save *millions* of lives. Yet, he is vilified by Republican members of congress who are more interested in political expediency and posturing than in the safety of the public, Some even attack the very concept of science, the scientific method. This is insanely dangerous. The undermining of the public's trust in the public health system actually killed people during the pandemic. I've complained about how some captains of industry treating their scientists, but the mistreatment of Dr Fauci by congressional Republicans tops everything.

To be clear: I'm not criticizing OpenAI's work nor their claims. I'm trying to correct a *perception* by the public & the media who see chatGPT as this incredibly new, innovative, & unique technological breakthrough that is far ahead of everyone else. It's just not.

I think you and Trump should do us a favor and launch yourselves to Mars tomorrow.

Wait, wat?

Oh your god.

You know what can *actually* be the downfall of Western civilization? People who dismiss facts, disparage scientists and the journalists, side with aspiring dictators, disseminate batshit-crazy conspiracy theories, and buy themselves a medium to amplify them.

Thank you IIT Madras for hosting me yesterday. I got to take a selfie with the audience of my Subaru Suresh Distinguished Lecture.

The year is 1440 and the Catholic Church has called for a 6 months moratorium on the use of the printing press and the movable type. Imagine what could happen if commoners get access to books! They could read the Bible for themselves and society would be destroyed.

I am extremely honored to be the recipient of the 2018 ACM A.M. Turing Award, and absolutely delighted to be sharing it with my friends and colleagues Geoffrey Hinton and Yoshua Bengio. NYT:... nytimes.com/2019/03/27/techn…

If you are a student or academic researcher and want to make progress towards human-level AI: >>>DO NOT WORK ON LLMs<<< LLMs are an off ramp. Thousands of engineers are working on LLMs with enormous computing resources. The only way you could possibly contribute is by analyzing existing LLMs and showing their power and limitations. But it's more fun and impactful to come up with new ideas and new architectures and show that they might work, even on small problems.

I'm a scientist, not a business or product person.

<sarcasm> Come work at ClosedAI. With AGI just around the corner, your shares will be worth 42 sextillionnollars. We can claw back your vested shares if you quit, unless you sign a non-disparagement agreement. Oh wait, sorry, we didn't realize our contract was this harsh until someone published an article on it. You can keep them. We're worth 42 octillionnollars, but since we're a private company, you can't really sell your shares. Also, we can actually stop you from selling them. And you can't say a word about what it is we're working on. If you do, we'll fire you within the hour, claw back your shares, and you won't be able to say *anything*. Don't take it personally but, you know, we're soooo far ahead of everyone else and AI is soooo dangerous in the hands of the unwashed masses. It's safe only if *we* do it. </sarcasm>

Dude, the Consumer Price Index has been back to a very normal 3% for a year now. Inflation was up between mid-2020 and mid-2022 because of COVID. Your tweet is total BS. bls.gov/charts/consumer-pric…

I don't wanna say "I told you so", but I told you so. Quote: "Ilya Sutskever, co-founder of AI labs Safe Superintelligence (SSI) and OpenAI, told Reuters recently that results from scaling up pre-training - the phase of training an AI model that uses a vast amount of unlabeled data to understand language patterns and structures - have plateaued." ... threads.net/@yannlecun/post/…

💥BOOM 💥 Llama 3.1 is out 💥 405B, 70B, 8B versions. Main takeaways: 1. 405B performance is on par with the best closed models. 2. Open/free weights and code, with a license that enables fine-tuning, distillation into other models, and deployment anywhere. 3. 128k context length, multi-lingual abilities, good code generation performance, complex reasoning abilities, tool use. 4. Llama Stack API for easy integration. 5. Ecosystem with over 25 partners, including AWS, NVIDIA, Databricks, Groq, Dell, Azure, and Google Cloud. Blog post: ai.meta.com/blog/meta-llama-… Llama home: llama.meta.com/

If you ever feel dumb, just remember there are some people who still believe everything shown in Elon's tweets.

Giving a talk at IIT Madras on October 22.

Altman, Hassabis, and Amodei are the ones doing massive corporate lobbying at the moment. They are the ones who are attempting to perform a regulatory capture of the AI industry. You, Geoff, and Yoshua are giving ammunition to those who are lobbying for a ban on open AI R&D. If your fear-mongering campaigns succeed, they will *inevitably* result in what you and I would identify as a catastrophe: a small number of companies will control AI. The vast majority of our academic colleagues are massively in favor of open AI R&D. Very few believe in the doomsday scenarios you have promoted. You, Yoshua, Geoff, and Stuart are the singular-but-vocal exceptions. like many, I very much support open AI platforms because I believe in a combination of forces: people's creativity, democracy, market forces, and product regulations. I also know that producing AI systems that are safe and under our control is possible. I've made concrete proposals to that effect. This will all drive people to do the Right Thing. You write as if AI is just happening, as if it were some natural phenomenon beyond our control. But it's not. It's making progress because of individual people that you and I know. We, and they, have agency in building the Right Things. Asking for regulation of R&D (as opposed to product deployment) implicitly assumes that these people and the organization they work for are incompetent, reckless, self-destructive, or evil. They are not. I have made lots of arguments that the doomsday scenarios you are so afraid of are preposterous. I'm not going to repeat them here. But the main point is that if powerful AI systems are driven by objectives (which include guardrails) they will be safe and controllable because *e* set those guardrails and objectives. (Current Auto-Regressive LLMs are not driven by objectives, so let's not extrapolate from their current weaknesses). Now about open source: your campaign is going to have the exact opposite effect of what you seek. In a future where AI systems are poised to constitute the repository of all human knowledge and culture, we *need* the platforms to be open source and freely available so that everyone can contribute to them. Openness is the only way to make AI platforms reflect the entirety of human knowledge and culture. This requires that contributions to those platforms be crowd-sourced, a bit like Wikipedia. That won't work unless the platforms are open. The alternative, which will *inevitably* happen if open source AI is regulated out of existence, is that a small number of companies from the West Coast of the US and China will control AI platform and hence control people's entire digital diet. What does that mean for democracy? What does that mean for cultural diversity? *THIS* is what keeps me up at night.

But seriously folks, this a short and juicy tirade in which I say: (0) there will be superhuman AI in the future (1) they will be under our control (2) they will not dominate us nor kill us (3) they will mediate all of our interactions with the digital world (4) hence, they will need to be open platforms so that everyone can contribute to training and tuning them.

Worth repeating: Do not confuse retrieval with reasoning. Do not confuse rote learning with understanding. Do not confuse accumulated knowledge with intelligence.

As I've pointed out a number of times, claims that immigrants vote illegally or that illegal immigrants are being let into the US to "import voters" are crazy-ass conspiracy theories and propaganda disseminated by people with a Trumpist political agenda, like @elonmusk. The statistics are clear: immigrants, legal or illegal, simply cannot and do not register to vote. They simply do not vote. The benefits are non existent and the risks are enormous (deportation). @elonmusk and I know this well: we're both immigrants.

Meta will *not* release the multimodal versions of its AI products and models in the EU because of an unpredictable regulatory environment. This means that EU users of Ray-Ban Meta won't be able to use the image understanding features. It also means that the EU industry will not have access to future multimodal versions of Llama-3. theverge.com/2024/7/18/24201…

Please ignore the deluge of complete nonsense about Q*. One of the main challenges to improve LLM reliability is to replace Auto-Regressive token prediction with planning. Pretty much every top lab (FAIR, DeepMind, OpenAI etc) is working on that and some have already published ideas and results. It is likely that Q* is OpenAI attempts at planning. They pretty much hired Noam Brown (of Libratus/poker and Cicero/Diplomacy fame) to work on that. [Note: I've been advocating for deep learning architecture capable of planning since 2016].

<not_an_april_fool_joke> By amplifying human intelligence, AI may cause a new Renaissance, perhaps a new phase of the Enlightenment. But prophecies of AI doom are also causing a new form of medieval obscurantism. </not_an_april_fool_joke>

Time to add folks who haven't left and are doing actual technical work, like Alec, Wojciech, and Aditya.

I don't wanna say 'I told you so', but I told you so.

Hard disagree. X merely tells you that someone, somewhere, noticed something. But then, one needs to know if this something is true or false, where it came from, what the context is, and whether it is, indeed, propaganda. Figuring this out is what good journalists do.

Many (rich) people on the Right do not actually believe in democracy. They have a huge superiority complex. They think the populace can be brainwashed into supporting policies that profit them and their rich friends at the expense of the majority. So they buy themselves brainwashing tools, such as a social network or a news corporation.

Elon Musk didn't just buy Twitter. He bought a propaganda machine to influence how you think. Rupert Murdoch didn't just buy Fox and the NY Post. He bought a propaganda machine to influence how you think. Sinclair didn't just buy 193 TV stations. They bought a propaganda machine to influence how you think. iHeartMedia/ClearChannel didn't just buy 855 radio stations. They bought a propaganda machine to influence how you think.

No. If it were the case, we would have AI systems that could teach themselves to drive a car in 20 hours of practice, like any 17 year-old. But we still don't have fully autonomous, reliable self-driving, even though we (you) have millions of hours of *labeled* training data.

Animals and humans get very smart very quickly with vastly smaller amounts of training data. My money is on new architectures that would learn as efficiently as animals and humans. Using more data (synthetic or not) is a temporary stopgap made necessary by the limitations of our current approaches.

Where are European companies?

Sometimes, the obvious must be studied so it can be asserted with full confidence: - LLMs can not answer questions whose answers are not in their training set in some form, - they can not solve problems they haven't been trained on, - they can not acquire new skills our knowledge without lots of human help, - they can not invent new things. Now, LLMs are merely a subset of AI techniques. Merely scaling up LLMs will *not* lead systems with these capabilities. There is little doubt AI systems will have these capabilities in the future. But until we have small prototypes of that, or at least some vague blueprint, bloviating about AI existential risk is like debating the sex of angels (or, as I've pointed out before, worrying about turbojet safety in 1920). bath.ac.uk/announcements/ai-…

Nope. It's perfectly legal. It's illegal for foreigners to contribute *money* to campaigns. But it's perfectly legal for foreigners to campaigns on behalf of a candidate Campaigning is speech. Speech Is protected by the 1st Amendment. As you probably know.

MMS: Massively Multilingual Speech. - Can do speech2text and text speech in 1100 languages. - Can recognize 4000 spoken languages. - Code and models available under the CC-BY-NC 4.0 license. - half the word error rate of Whisper. Code+Models: github.com/facebookresearch/… Paper: scontent-lga3-2.xx.fbcdn.net… Blog: ai.facebook.com/blog/multili…

Today, I was made a Chevalier de la Légion d'Honneur by President Macron at the Élysée Palace. Président Macron gave a wonderful speech.

You know what's unethical? Scaring people with made-up risks of a technology that is both useful and beneficial.

"Our country is being poisoned" by immigrants, like me, @elonmusk, @sundarpichai, @satyanadella, @vkhosla, Jensen Huang, and two of Trump's three wives.

Language is an imperfect, incomplete, and low-bandwidth serialization protocol for the internal data structures we call thoughts.

So, you've gone full-throttle sexist now? This "observation," which is not based on any actual observation, flies in the face of every study of conservatism in social psychology. Many studies show that conservatism is associated with social conformity and a submission to group-based dominance structures such as paternalism, religion, military discipline, etc. scholar.google.com/citations… In baboons, social status is largely based on individual physical strength. But in humans, it is largely based on prestige, which results from intellectual influence and value to the group. It leads to strong groups. Evolutionary psychology has a name for the dominance-prestige duality: the Dual Strategy Theory of social hierarchy. en.wikipedia.org/wiki/Dual_s… Speaking of low T and physical vulnerability: we're still waiting for that cage fight. In the meantime, keep talking to baboons. PS: The good thing about superhuman AI is that it won't have testosterone. Whatever influence it will have will be because of positive intellectual effects on society. Not because of a desire to dominate.

Do you actually believe this crap? Either you do, and one may be entitled to question your sanity. Or you don't, which means you're deliberately lying and have a very low opinion of your followers' intellects.

New paper: turns out you can train deep nets without normalization layers by replacing them with a parameterized tanh()

Yes, save AI-generated pets!

You are undermining democracy by spewing crazy-ass conspiracy theories that erode the public's trust in democratic institutions. Non-citizens can not and do not vote.

CoPE: Contextual Positional Encoding. A new paper from FAIR that @elonmusk could use to improve Grok.

No. If it were the case, we would have AI systems that could teach themselves to drive a car in 20 hours of practice, like any 17 year-old. But we still don't have fully autonomous, reliable self-driving, even though we (you) have millions of hours of *labeled* training data.

Complete bullshit. Pardon my french.

Let me clear a *huge* misunderstanding here. The generation of mostly realistic-looking videos from prompts *does not* indicate that a system understands the physical world. Generation is very different from causal prediction from a world model. The space of plausible videos is very large, and a video generation system merely needs to produce *one* sample to succeed. The space of plausible continuations of a real video is *much* smaller, and generating a representative chunk of those is a much harder task, particularly when conditioned on an action. Furthermore, generating those continuations would be not only expensive but totally pointless. It's much more desirable to generate *abstract representations* of those continuations that eliminate details in the scene that are irrelevant to any action we might want to take. That is the whole point behind the JEPA (Joint Embedding Predictive Architecture), which is *not generative* and makes predictions in representation space. Our work on VICReg, I-JEPA, V-JEPA, and the works of others show that Joint Embedding architectures produce much better representations of visual inputs than generative architectures that reconstruct pixels (such as Variational AE, Masked AE, Denoising AE, etc). When using the learned representations as inputs to a supervised head trained on downstream tasks (without fine tuning the backbone), Joint Embedding beats generative. See the results table from the V-JEPA blog post or paper: ai.meta.com/blog/v-jepa-yann…

A new family of antibiotics discovered with graph deep learning.

Suddenly, many of your tweets look like jokes, Pedro.

Calm down, dude. We will become a multiplanet species regardless of who gets elected. We will even become a multiplanet species if you die tomorrow. Our multiplanet future will happen faster if Kamala Harris is elected because you need someone who: (1) will not crash the economy, (2) will fund science, technology, research, and education, (3) won't claim that scientists should be sent to prison, (4) won't dismiss universities as Marxist, woke, or whatever.

World models, intuitive physics, planning, problem solving, discrete search for solutions, continuous optimization of control parameters... Dogs manage to do all this with 2 billion neurons. Why debate human-level AI when we can't approach dog-level intelligence yet? 1/N

This item is news. News you don't like, but factual news. I realize you may have lost touch with the concept of factual news. Google is not controlled by "far left activists" and neither is Meta. But X is objectively controlled by a conspiratorial, far right propagandist.

People studying misinformation lean left for two reasons: 1. scientists lean left, regardless of specialty, because they care about facts. 2. misinformation today primarily comes from the Right ("they're eating the dawwwgs!") which makes it worth studying and fighting against for people leaning left.

Dear journalists, it makes absolutely no sense to write: "PaLM 2 is trained on about 340 billion parameters. By comparison, GPT-4 is rumored to be trained on a massive dataset of 1.8 trillion parameters." It would make more sense to write: "PaLM 2 possesses about 340 billion parameters and is trained on a dataset of 2 billion tokens (or words). By comparison, GPT-4 is rumored to possess a massive 1.8 trillion parameters trained on untold trillions of tokens." Parameters are coefficients inside the model that are adjusted by the training procedure. The dataset is what you train the model on. Language models are trained with tokens that are subword units (e.g. prefix, root, suffix). Saying "trained a dataset of X billion parameters" reveals that you have absolutely no understanding of what you're talking about.

AI is not some sort of natural phenomenon that will just emerge and become dangerous. *WE* design it and *WE* build it. I can imagine thousands of scenarios where a turbojet goes terribly wrong. Yet we managed to make turbojets insanely reliable before deploying them widely. The question is similar for AI: "do we think there exists at least one design of an AI system that is simultaneously safe/controllable, and can fulfill objectives in more intelligent ways than humans ?" If the answer is yes, we'll be fine. If the answer is no, we won't build it. Right now, we don't even have a hint of a design of a human-level intelligent system. So it's too early to worry about it. And it's way too early to regulate it to prevent "existential risk."

Almost all of the federal funding to universities goes to STEM and biomedical research. The amount going to social science is a rounding error, and to humanities (including social criticism), it's essentially zero. STEM research in universities is the root of *all* the technologies your companies are based on. Do you really want to kill the golden goose? Note: HHS funding mostly goes to medical research and life science through NIH. Cutting those funds wouldn't just kill the golden goose. It would kill people.

Simply put, Trump is a traitor.

Hey @elonmusk, this is your champion of free speech? Putting people in jail for criticizing Supreme Court justices?

Some folks say "I'm scared of AGI" Are they scared of flying? No! Not because airplanes can't crash. But because engineers have made airliners very safe. Why would AI be any different? Why should AI engineers be more scared of AI than aircraft engineers were scared of flying?

If language were sufficient to express human thought, why would we need visual arts, music, dance?

I testified in front of the US Senate Subcommittee on Intelligences yesterday. I argued that the rapid and free exchange of ideas, scientific publications, open source code, and trained models is the reason that AI has progressed so fast in the last decade. I also pointed out that this openness is what keeps the US ahead. intelligence.senate.gov/hear…

I actually do have Elon's number. But we haven't spoken in a while.