@RichardSocher
Building self-improving superintelligence CEO @recursive_si and @youdotcom MP @aixventuresHQ Ex: Stanford Adj Prof, Chief Scientist at Salesforce, CEO MetaMind
I don't know man, BitWhisper required those CPUs to be 0-40cm apart and had a bitrate of 8bits per hour. I think calling people idiots who disagree with you isn't helpful but it's X so I get it. I think one difference is that their creativity doesnt stop with the attacker but continuous to the defender which is arguably more work but leads to more realistic scenarios.
roon@tszzl·everyone who understands the first thing about computer security or what superintelligence means understands this is possible and all the usual gang of idiots is calling this scifi hype x.com/firesidealpha/…
This was a great conversation on many of the important topics of today re safety etc.
Latent.Space@latentspacepod·🆕 Humanity’s Last Invention — with @RichardSocher! t.co/ZSeHSpcC1g @recursive_si is the latest neolab to burst on the scene with a $5B fundraise and one of the most impressive cofounder lists ever assembled to tackle open ended, self-improving AI for AI research. We dive into the Eureka Machine, Richard's 10 Spaces of Intelligence (so much left until AGI!), DecaNLP vs @alecrad and why AI peer review is broken, and why this may just be the last invention that humanity has to do on our own. Timestamps 00:00:00 The Eureka Machine and Superintelligence 00:02:23 AI Optimism, Slow Takeoff, and Regulation 00:07:56 AI Safety, Reward Hacking, and Anthropic’s Constitution 00:11:49 Alignment, Personalization, and Open Source AI 00:15:46 Why Richard Started Recursive 00:20:03 Recursive Self-Improvement and the Founding Team 00:22:55 Are Today’s LLMs Enough? 00:29:03 DecaNLP, GPT, and the Rejected Idea Ahead of Its Time 00:34:38 Open-Endedness and Evolutionary AI 00:36:38 What Happens When AI Chooses Its Own Goals? 00:41:16 Superintelligence for Science 00:42:40 GPUs, Compute, and the Limits of AI Takeoff 00:45:07 Recursive’s Results: AI Beating Humans and Their Agents 00:49:14 Reward Engineering and Auto Research 00:53:12 The AI Economist and Simulating Entire Economies 00:58:07 LLM Simulations, Personas, and Mode Collapse 01:03:38 Recursive’s Roadmap, Agents, Search, and Finance 01:09:13 The Upper Bounds and Spaces of Intelligence 01:30:21 Goals, High Agency, and Advice for Builders
This is what I describe in great detail in my upcoming book The Eureka Machine. AI will help us massively accelerate science. Which is not zero sum and will progress humanity. Love seeing it play out in real time.
Arthur Turrell@arthurturrell·Today @Google, we published work led by @m_codreanu on AI in Science, drawing on 15 million Gemini interactions, 2,600 specialised AI models & a 600-scientist survey. What did we find about how AI is changing science? Read on...
There is NO realistic scenario where AI wipes out all of humanity. I've talked to many AI safety experts about existential risks. Some of the "experts" are just phenomenal sci-fi authors. There are also real researchers working on realistic risks. They point out real and likely harms and I'm explicitly not arguing that there could be no such harm. We need to take these seriously. However, whenever I ask for an actual detailed scenario that would kill every single human i.e. extinction, it usually starts and ends with "just look at current progress and use your imagination". No science, no scenario planning, just handwaving probabilities without data, based on gut-feelings, often little expertise, sometimes real expertise combined with a huge amount of technological optimism turned into fear and then trying to quantify it post hoc. When pressed further the conversations often go into different directions: 1) Completely crazy and cool sci-fi story telling: grey blob, time travel, attacks in the 15th dimension that end the planet in 12 seconds, etc. Elizier level stuff. Remember, the Terminator movie started with TIME TRAVEL. Fun movie plot, but not necessary to engage with further. 2) In the rare cases where I was actually given a more detailed scenario, it still required near magical capabilities that are outside what we know to be possible with the laws of physics or biology. Hard take off scenarios that forget how hard it is to procure materials and machines, get human attention, etc.. What many of these scenarios assume is that attackers have near magical capabilities but defense stays still. Attackers will have the most sophisticated cyber weapons but defenders just sit still, no security patches, no red teaming, etc. We already see it going the other way, Rainbow teaming has made LLMs safer, major companies like Microsoft have more security patches than ever before thanks to AI. OpenAI thinks they solved all P0 security issues thanks to AI. Cyber security will continue to be a cat and mouse game, but both sides have more powerful tools. In fact, I lay out in another post how the asymmetry may eventually even flip thanks to AI. Bio risk scenarios I think are the most serious and worthy of debate. But even there, the scenarios to "wipe out all of humanity" go something like: Magical viruses with perfect transfer, perfectly undetectable over years of spreading, zero side effects, then with an off switch that's somehow remotely executable, perfectly automatable lab, no leaks, super sophisticated machines bought and nobody notices the lab, etc. But somehow nobody on team humanity has such sophisticated knowledge to save us and hence nobody builds a magical super vaccine. This is a good read on the bio risk subject: . Realistically, one can already build viruses. There’s a reason gain of function research has been outlawed. However, we also know that as viruses spread they often become less deadly in order to keep spreading. That’s how the Spanish flu became the seasonal flu. Current knowledge of biology doesn’t have remote controlled on/off switches. Again, it’s possible there’s a lot of harm, but it’s unrealistic to assume this wipes out all of humanity. It’s at this point of my discussions that we usually can come to an agreement that even though a virus is unlikely to wipe out all of humanity, it’s already bad if any single person or a large group of people dies. 3) The AI will somehow be able to convince all humans to vote against their own self interest and nobody will notice until we're dead. Because it is so smart, it somehow has this world wide reach to manipulate everyone. Like, it just solves all of marketing and has an infinite budget. It's sort of like saying the highest IQ people always have the most followers and everybody listens to them... Hasn't happened and is unlikely to. Sure we can manipulate people into clicking on engaging short form content and wasting many hours, but try getting people to do their math homework or something truly hard, it's not that easy. Again, I’m just arguing against the likelihood of human extinction. I can totally see an Idiocracy or WALL E type scenario where people stop using their brains and education doesn’t keep up and people get lazier. Totally suboptimal but not the end of human life, certainly not by the end of the decade. It would take many generations and we should update education to prevent this. Let’s start gyms for the mind. – Like electricity, AI will be everywhere and needs to have safety guardrails. In fact, the history of electricity has many parallels - two companies fighting over whose electricity is safer and better, demonstrating how dangerous it is by electrocuting animals / training agent swarms to hack into systems. Of course, electricity is a necessary ingredient to AI, but AI can be given more agency so history will not repeat but only rhyme here. Remember, the benchmark OpenAI's swarm was trying to solve was called ExploitGym, a cybersecurity evaluation consisting of hundreds of capture-the-flag style puzzles designed to measure *offensive cyber capabilities*. Well, maybe if you play that game, you will win that prize. Should we allow such misaligned, reward-hacking swarms to explicitly work on offensive hacking? I don’t think so. Reasonable regulation will regulate AI applications in each industry. For example, you may want to put some generative AI behind a rating if it creates visual or textual content that's not safe for children, like we do with movies. We should continue to be very careful with how we train biological AI and only collaborate with responsible biotech companies. We keep gain of function research illegal. We should only let self-driving cars on the street after significant safety testing. etc. We need to work on those real issues and continue to be vigilant in how we develop and deploy this technology. At Recursive we're spending a lot of resources on researching reward hacking and alignment. We're also not working on replacing jobs but on creating novel knowledge and automating the scientific method. Our agent swarms invent, implement, validate, and criticize each other’s ideas, among many other things. The goal of increasing knowledge is not zero sum and has real potential to improve humanity. I also think such rewards and environments are much more aligned with future AIs that want to see humanity and our creation of knowledge thrive. Ultimately, capitalism will be a very useful defense mechanism against rogue AI. Note that no agent at OpenAI truly "escaped" in the sense that it actually now runs on some other servers. Why? It’s not only because Huggingface doens’t have the same massive compute cluster thatn OpenAI has. It is also because that would be an insanely expensive IP loss for a for-profit company. So there are likely more safeguards in place to prevent complete model exfiltration. Rogue AIs running on billion dollar data centers without anybody noticing? Unlikely because companies will turn that off immediately and clean it up because it would be an insane loss otherwise. An AI that can truly move beyond objective functions and choose its own subjective functions (and then somehow decides to kill us all instead of going to explore the universe)? Nobody is working on that because when a company spends billions of dollars to have an AI work for you or answer your emails, it doesn't want it to go "Meh, your emails are boring, I'd rather explore the molecular composition of the atmosphere on Venus, bye." Of course, there might be suboptimal sub-goals and we do need to improve human reward engineering and reduce our reliance on RL as the only path to alignment. Extraordinary claims, require extraordinary evidence. Certainly some safety researchers have real data and benchmarks. Many truly think carefully about specific failure cases and risks.. However, made up precision like 20% chance that humanity dies within the decade (ie in 4 years!) is just a weird psychosis that's filling a void for many people who want to believe in the apocalypse again and the four horsemen have lost their appeal. Some people who purport this narrative might do it for money or fame. I am sure some really believe it also. They forget how easily those narratives can be turned against them. When people hear there’s a 1% chance of infinite death and the end of humanity, the life improvements and cancer cures and new battery materials AI can invent don’t matter much anymore. Bernie Sanders now proposes 10 years of prison for folks working on RSI. That should give some of the researchers real pause who talk up these scenarios and then go right back to pushing the frontier. I do agree that companies that are accidentally competing on "felony bench" (ie how many felony hacks their AI can do), might want to consider pacing themselves. Open source won't pace, China won't pace, anybody else working to catch up will not pace. I hope the folks actually working on AI and TechBio, curing diseases, speeding up FDA approvals with organoids and AI, making self driving safer, and all the other wonderful applications of AI, I hope they won't pace either. The one exception that may be considered borderline realistic and would be a path to almost complete annihilation would be to connect the only tool we built for such a purpose - nuclear weapons - to the internet and give AI full control over it without human oversight. AI here is as dangerous - as would be random number generator. The nuclear weapons are the core danger. Giving AI such access would be incredibly stupid and careless. We must avoid it at all costs. Fortunately, nobody is advocating for this. Even the Terminator 3 movie used it as the starting point of Skynet. Generally, we must avoid making kill decisions by AI as much as possible. People are already up in arms about one hacking failure that afaics has not caused financial harm and so nobody got sued over it. Imagine how much people would really be up in arms if AI had caused real harm. If we made companies liable for the felonies their AI commits, these problems would get solved very quickly. Generally, I believe that people will continue to update their beliefs and adjust their laws according to risks. The EU already regulates large models and hence hurts its fledgling AI ecosystem. Other countries will follow when the risks get more real. Long-term, no government whether democratic or authoritarian wants to lose control to an AI so the complete loss of control in the political domain will remain extremely unlikely. The best way to understand why we must not regulate abstract AI models (rather than their real applications) is to drop the A in AI: How do you regulate intelligence? Should smarter people just go to prison because they could cause more damage? My hypothesis here is that such abstract AI regulation would require an unprecedented level of totalitarian control and surveillance. You'd need to know what everybody asks the model on their local GPU. You would literally have to start an international thought police and make sure that all ideas, conversations and types of intelligence are compliant. The risk of losing our freedoms is much more real than the existential risks purported by the extreme safety experts. If anybody thinks they can actually lay out a realistic scenario, I'd be happy to debate them again. To be clear, this is just a weekend post that got a bit too long. It is my personal opinion and not the official position of any company. We have a diverse set of opinions within Recursive on this subject. We certainly all agree that bad outcomes are possible with AI and we should continue our work on preventing those.
David Bellamy@DavidRBellamy·I must be among an extremely small group of people (n=1?) that have both 1) trained a frontier LLM and 2) designed and synthesized custom viruses in a lab with my own two hands. And I think that the takes on AI killing us all by creating dangerous viruses is total bogus.
This is a useful distinction here also:
Dan Elton@moreisdifferent·Many don't seem to fully grok how much worse extinction is vs catastrophic risk. It makes sense EAs are focused on extinction risks. I asked ChatGPT to make a diagram. And BTW, AI is likely not conscious and there's a good chance we are the only intelligent life in our galaxy.
Really important and timely conversation around the safety of recursive self-improving superintelligence. Extraordinary claims of extinction probabilities require extraordinary evidence not extrapolation plus handwaving about progress. Thanks @nxthompson for asking the hard and interesting questions. t.co/7RFo6MBgbU
We can safely advance science and society with AI or get distracted with scifi doom scenarios that require magical extrapolation. I really enjoyed this podcast on RSI for science. youtube.com/watch?v=kyyLku…
I think this take on AI and security is correct. Less doom. More AI optimism is possible wrt jobs security etc. Wrote about this AI safety paradox here: socher.org/thoughts/ai-sa…
Perry E. Metzger@perrymetzger·Yesterday, Microsoft's monthly Patch Tuesday had fixes for 974 security vulnerabilities, almost all of them found by AI systems. That's a ridiculously large number, a new record by far in fact. Does this mean we're seeing some sort of AI security apocalypse? No, quite the opposite. It means that we're finally clearing out the vast number of security holes that have been lurking all this time in our software. The Doomer view is that this will continue without end, and that if you keep getting smarter AI systems they will *always* find new bugs. That's simply untrue; it implies that all software has an *infinite* number of security holes, but a program with a finite number of lines of code simply cannot have an infinite number of vulnerabilities. What we actually have is a large but limited pool of problems, and the AI systems are rapidly finding them. Eventually, and eventually isn't that far off, the well is going to start drying up. It will get harder and harder to find new security holes. Over the next few years, we will also start doing formal verification of software, that is, mathematically proving that the software lacks bugs of certain sorts. (AIs turn out to be very good at formally proving things.) So, what's happening is *good*. We are rapidly finding bugs that have been lurking for years and sometimes decades, and we're removing them, and newly built software will get AI examination and will be much less likely to have security vulnerabilities in the first place. The situation is getting better, not worse, and it's getting better rapidly. We have been in a continuous computer security crisis since the Morris Worm in 1988. We are finally starting to climb out of it, thanks to AI. This is not a tragedy at all.
It is a travesty how much less celebrated the stats of NLP are compared to those of basketball. NLP > NBA ... in terms of relevance for humanity..
Matthew Graham@mattyryze·China is putting scientists on public screens and treating them like national celebrities, we are so fucked
"I found this book to be quite useful in pondering the big questions about our shared frontier." Thank you @johnkwerner for this thoughtful review of The Eureka Machine in Forbes. Out September 22. forbes.com/sites/johnwern…
Link to preorder amazon.com/gp/product/154…
Incredible result.. With better search the different between models becomes much much smaller!
Braintrust@braintrust·Imagine doing your job without ever looking anything up on the internet. That's an agent without web search. Giving agents the web changed what they can do, especially on anything recent that isn't reflected in training data. But agents don't search like people, so optimizing their performance is a new challenge. There's a lot of great research out there on search behavior, but we wanted to answer a more operational question: when should search be on, and how should you configure it? We evaluated 1,329 current events questions across 4 models and 14 conditions, comparing @youdotcom, provider built-in search, and no search. We found that: - Search reduced the gap between models from 47.9 points to 5.6 - Retrieval gain declined with event age, from ~45 points for recent events to ~24 for the oldest - Runs with 5+ searches scored 19–49%. A fifth query was associated with lower performance Read the research → t.co/M7KB7Lljhi
I have talked about this and several other laws (like taxing unrealized gains for founder shares) to Chancellors Merkel, Scholz and Merz and many others in their cabinets. Some improvements have happened in the last years but not enough ... I'll keep trying. Germany could unlock a lot of entrepreneurial energy. Maybe it needs some special economic zones like China had back in the day?
Patrick Collison@patrickc·Met a German founder this week and asked him if all the stories one reads about the challenges of startups in Germany are exaggerated. "No, they're understated." Proceeded to describe spending a full day having a 90-page investment contract read to him (mandatory under German law; § 13 BeurkG) by a notary that then charged €30,000. That was for his first company. His second company, needless to say, was not incorporated in Germany.
If accuracy and time matters to your agent. You got it.
It's sad to observe this unfold in slow motion... The EU regulating and slowing down innovation with one sort of sensible regulation after another.
Perry E. Metzger@perrymetzger·A fascinating little glimpse at how Europe keeps destroying its entrepreneurs, one "tiny little obviously reasonable who could possibly object" regulation at a time. The US has a thriving ecosystem of small electronics manufacturers; Europe has just killed its equivalent companies for good. This is just one of thousands of similar regulations that have largely eliminated European technology startups. t.co/zAHXAGxNxd
Some of the scientists I respect don't think AI will help cure cancer or solve our biggest problems. I disagree. After nearly 20 years working in AI and NLP, I could not be more hopeful about the future. I believe that accelerating scientific discovery will be the most meaningful application of AI. Science was never limited by ambition, it was limited by people. There will never be enough scientists to work on every problem worth solving. AI will change that. I've always said the future needs better marketing. I spent the last couple of years laying out the more likely and quite positive vision for AI and our future in my upcoming book. It's called The Eureka Machine, and it comes out September 22.
Link to preorder: amazon.com/gp/product/154…
How to balance 1. Incrementally fixing the misery of the moment; and 2. Long-term advances in science and technology to truly solve the underlying problems; is the most important philosophical question of humanity. It cuts across the political spectrum. Access to novel tech and scientific advances always starts with a select few, then scales. At full scale, we end up in a place where billionaires and an average teenager have a similar smartphone, a similar warm water shower, the same antibiotics, the same knowledge access, etc.
Jesús Fernández-Villaverde@JesusFerna7026·How bad was life before modern science, technology, and economic growth? Awful. The example I will use in class this Wednesday is Henry Clay (1777-1852). He was one of the most powerful men of his time in the U.S., not just among the elite but among the elite of the elite: Speaker of the House, Secretary of State, Senator, and three-time presidential candidate. He was also rich and a large slaveholder. The U.S. was among the richest countries in the world (perhaps only behind Great Britain) and among the most technologically advanced. And we are not talking about the early Middle Ages, but about the first half of the 19th century. Given the constraints of his time, he had nearly everything within reach. Clay and his wife, Lucretia (both in the photograph), had eleven children: six daughters and five sons. He buried all six daughters. By 1835, all were dead: two as infants, two as children, two as young mothers. He also buried a son in a war his father had opposed. Another son, Theodore, fractured his skull as a child, but since nobody could treat him, he spent decades in an asylum in Lexington. Think about this for a second: seven children dead and one handicapped out of eleven, in the first half of the 19th century, among the elite of the elite, in one of the richest and most advanced places on the planet. And I was angry yesterday because my cell phone coverage in Toddington (England) was poor.
How to balance 1. Incrementally fixing the misery of the moment; and 2. Long-term advances in science and technology to truly solve the underlying problems; is the most important philosophical question of humanity. It cuts across the political spectrum. Access to novel tech and scientific advances always starts with a select few, then scales. At full scale, we end up in a place where billionaires and an average teenager have a similar smartphone, the same warm shower, the same antibiotics, the same knowledge access, etc.
Jesús Fernández-Villaverde@JesusFerna7026·How bad was life before modern science, technology, and economic growth? Awful. The example I will use in class this Wednesday is Henry Clay (1777-1852). He was one of the most powerful men of his time in the U.S., not just among the elite but among the elite of the elite: Speaker of the House, Secretary of State, Senator, and three-time presidential candidate. He was also rich and a large slaveholder. The U.S. was among the richest countries in the world (perhaps only behind Great Britain) and among the most technologically advanced. And we are not talking about the early Middle Ages, but about the first half of the 19th century. Given the constraints of his time, he had nearly everything within reach. Clay and his wife, Lucretia (both in the photograph), had eleven children: six daughters and five sons. He buried all six daughters. By 1835, all were dead: two as infants, two as children, two as young mothers. He also buried a son in a war his father had opposed. Another son, Theodore, fractured his skull as a child, but since nobody could treat him, he spent decades in an asylum in Lexington. Think about this for a second: seven children dead and one handicapped out of eleven, in the first half of the 19th century, among the elite of the elite, in one of the richest and most advanced places on the planet. And I was angry yesterday because my cell phone coverage in Toddington (England) was poor.
Always super inspiring to catch up with Peter! x.com/PeterDiamandis…
We just demoed how an agent can get real-time financial research through the @youdotcom Finance Research API and pay for it automatically, with no person setting up API keys, accounts, or payment info ahead of time via the @coinbase x402 standard. An agentic trader needs up-to-date research (e.g., SpaceX stock news) to make informed trading decisions. To solve, configure a cron job to run every trading day at 9:28 AM (two minutes before market open). The agent: 1. Requests a financial research report from t.co/MaOM1Mwww8 using x402 as payment 2. Approves the micro-transaction from a crypto wallet (configured via MCP for any wallet provider) 3. Receives the research instantly 4. Executes up to 3 autonomous trades, with a $250 cap per trade No API key, no credit card, no KYC form. Agents pay for t.co/MaOM1Mwww8's APIs directly, in crypto. That's one of the first real production use cases for crypto micropayments. Read more here: t.co/yRZcwbvijY
I think it's important to explain to lay people that when the OpenAI model hacked Huggingface they did not "escape" or "break out" in the sense that its actual computation could now be run elsewhere. It's not like a prisoner that actually escaped and is now outside. It's more like a prisoner who was able to fly a drone outside of prison but they themselves are still very much inside. The model could still be very easily turned off and I'm pretty sure OpenAI makes sure that its models would not actually replicate outside in terms of their entire model weights and code, etc. Because that would be a massive multi billion dollar loss vs a simple cyber hacking charge.
Case in point for stupid disinformation to stoke Ai fear.
Happy to welcome Laura Moss as CFO of @Recursive_SI Building frontier AI takes world-class research & engineering but also smart capital management. We understand that the sums we are dealing with are very large for startups, and we take that responsibility very seriously. We're thrilled to have Laura's expertise on our team. Laura served as SVP Finance at Cohere for over three years. Before that, she spent over 7 years at Google in roles of increasing responsibility across Technical Infrastructure and YouTube finance, supporting the business during a high growth and investment period. She holds an MBA from Wharton and a BSE in Biomedical Engineering from Duke. Welcome to the team, Laura! Excited to partner as we build something incredible together.
Anybody remember this whole marketing spiel about Constitutional AI and how it's supposedly the safer path to AGI? ... Screenshot from the current website. ... That approach is clearly not working
"Therefore, unlike the nuanced cost-benefit analysis that governs most of Claude’s decisions, these are non-negotiable and cannot be unlocked by any operator or user. Because they are absolute, hard constraints function differently from other priorities discussed in this document."
Building this Eureka Machine is my life's goal. It will automate scientific discovery for humanity which is also the most meaningful application of superintelligence. x.com/aiDotEngineer/…
How do we get people to go to gyms of the mind after recursive superintelligence gets here? Just like a normal gym, it's not that useful but it's good for you.
When it comes to copyright and AI capabilites there's a growing divide between the EU - US - China. Especially Germany and the EU hate how the sausage is made but will eventually enjoy it regardless. Chinese open source model builders have access to all ideas, copyrighted texts, comic book and anime characters, etc. Unrestricted flow of content and ideas is an advantage and - all things being equal- will result in superior models. This creates a tricky situation for the western world. Stick to valid and valuable and morally good copyright and ip laws. Let China overtake and build the best AI models. Then become completely dependent on external AI models. No simple clean answers here that all stakeholders will like.
There are at least a few reasons for why the incredible progress in AI hasn't yet resulted in a massive increase in GDP (some from Captain Obvious but number 3 is less intuitive to many smart people). 1. AI replaces some steps in complicated processes but companies are still doing mostly similar things and adoption and rethinking entire industries are slow. 2. Startups that replace everything (eg AI native law firm that is much cheaper) still need to ramp GTM, sales, etc But more importantly and surprising to many in Silicon Valley: 3. A huge chunk of the economy just does not require that much intelligence and won't materially change at its core with intelligence being abundant and cheap, eg. - tourism - people will want to see the pyramids with or without AI, - real estate - people want to live in hip and safe neighborhoods, AirBnB, rentals, etc. - luxury goods and status symbol bs, eg fancy handbags, clothes, overpriced cars, etc - food and large parts of the food supply chain (yes, I love AI for agriculture but crops and cows still need time to grow, etc) - sports and much of entertainment - oil drilling, tree growing/logging for construction, most of mining - etc If your existing economy depends mostly on these types of industries, AI won't impact it that much. But there's a whole new economy of knowledge work, research heavy industries, deep tech, online and digital work etc that will massively benefit and outgrow these existing industries.
As more people start working with AI the bar for what's an acceptable work product (slides, spreadsheet, documents, etc) will increase. It will be even less acceptable to present poorly thought through results. People who present sub-AI quality will have a hard keeping their job. It just hasn't happened broadly yet because most managers still don't know how to use AI well. Just like in chess the top people will get even better.
Adam is always wonderful to talk to! AI for science is our goal @Recursive_SI It's also what my upcoming book The Eureka Machine is about. It lays out how to build the ultimate invention generating machine for physics but also chemistry, biology, neuroscience, economics and astrophysics
Prediction: We will reinvent salary bands and management hierarchy for AI. "Junior" models have cheaper hourly/token costs, they do a lot of grunt work, some busy work, more low impact decisions. Senior frontier models will synthesize all the facts, make important decisions and delegate to those junior models. People will become managers and have their own agent org. Recursive will be well positioned for this future.
Yes, education needs to change because of AI. But just because AI can now do something better doesn't mean people shouldn't learn that skill. Not every skill we learn has to be immediately useful in a job. A lot of the training in school and university is around how to be a functional member of that society, how to learn anything (meta-learning), how to memorize anything, how to think and debate on the fly, how to communicate intentions well. Grades are a signal that can be useful in hiring (there are many others). Not only to know that a student understood that particular skill but also how they compare to their peers, if they can deliver under pressure, do they have the will to succeed. By taking many subjects students may get lucky and find something they're truly interested in. Sure, AI knows any fact and has lots of skills that may take you a long time to learn but you'll not be a very interesting conversation partner if you have to constantly get the next most interesting fact or a clever follow-up or counter from an LLM. You'll be easily fooled if you can't do basic math in your head. Ultimately, a lot of education may feel like a gym for the mind. Going to the gym isn't useful for humanity and mostly doesn't earn you more money. But it's good for you. It's good to use your muscles and your brain to not waste away. What we do need is teaching more agency, creativity and how to clearly communicate intentions and rewards to AI.
