"Intelligence is largely a power of prediction (see ARC, IQ tests, NLL). What is the pattern? What is the next figure after these? AI learns to count AI reasons on fifth grade level AI reasons on Ph.D level [?] A. AI answers fundamental questions B. AI turns your dog to paperclips" https://x.com/teortaxesTex/status/1811625585510170628?t=88PGdwrce7AL9yJ6RBE4QQ&s=19 https://phys.org/news/2024-07-artificial-intelligence-quantum-reality.html [Scientists Discover a Cause of Lupus and a Possible Way to Reverse It - News Center](https://news.feinberg.northwestern.edu/2024/07/10/scientists-discover-a-cause-of-lupus-and-a-possible-way-to-reverse-it/) [Ex-Meta scientists debut gigantic AI protein design model](https://www.nature.com/articles/d41586-024-02214-x) https://www.reddit.com/r/AskPhysics/s/58hB3yRQiR https://arxiv.org/abs/2402.09090 https://arxiv.org/abs/2407.06677 [Together MoA — collective intelligence of open-source models pushing the frontier of LLM capabilities](https://www.together.ai/blog/together-moa) [Road to AGI: The Importance of Clear Objectives and Reward Functions[v1] | Preprints.org](https://www.preprints.org/manuscript/202407.0687/v1) Neurons under microscope video https://x.com/ChanZuckerberg/status/1811100814141759993?t=LmWanwhFaSaaLg8-4JNZqQ&s=19 "Lately, I think of AI (mechanistic) interpretability research demos in terms of 5 tiers of rigor: 1. Pontification 2. Basic Science 3. Streetlight/Toy Demos 4. Useful Engineering 5. Net Safety Benefit" 6. https://x.com/StephenLCasper/status/1811536196562223161 https://arxiv.org/abs/2402.06908 [Architecture of the Human Default Mode Network: cytoarchitecture, wiring and signal flow | bioRxiv](https://www.biorxiv.org/content/10.1101/2021.11.22.469533v3.abstract?%3Fcollection=) [Fast Fourier transform - Wikipedia](https://en.wikipedia.org/wiki/Fast_Fourier_transform) [AI as a Model for the Brain | SpringerLink](https://link.springer.com/chapter/10.1007/978-3-662-68980-6_21) This is the biggest playing of God out there with manipulating the number of heads that an organism grows by Michael Levin! Two-headed worms 'coded': Bioelectric code behind worm self-healing manipulated Altering bioelectric networks reveals a 'hidden code' that controls the number of heads produced after injury even in normal-appearing flatworms Long-Term, Stochastic Editing of Regenerative Anatomy via Targeting Endogenous Bioelectric Gradients "Temporary modulation of regenerative bioelectric dynamics in amputated trunk fragments of planaria stochastically results in a constant ratio of regenerates with two heads to regenerates with normal morphology." https://x.com/burny_tech/status/1811842517161378256 https://www.sciencedaily.com/releases/2017/05/170523124219.htm https://www.cell.com/biophysj/fulltext/S0006-3495(17)30427-7?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS0006349517304277%3Fshowall%3Dtrue https://www.cell.com/cms/10.1016/j.bpj.2017.04.011/attachment/8c8eff76-be14-4f31-aca2-b1add8d0a329/mmc1 https://chrisfieldsresearch.com/GJ-memory-model.htm [What can worms and frogs teach us about collective intelligence? - YouTube](https://www.youtube.com/watch?v=Bdc-KQn-bwo) [Unveiling the Mind-Blowing Biotech of Regeneration: Michael Levin - YouTube](https://www.youtube.com/watch?v=Z0TNfysTazc) [Associative Memory in Hopfield Networks Designed to Solve Propositional Satisfiability Problems - YouTube](https://www.youtube.com/watch?v=r3onu55UL6E) Minimal cognitive intelligence and agency toolbox is goal directed behavior and control of internal states [Unveiling the Mind-Blowing Biotech of Regeneration: Michael Levin - YouTube](https://youtu.be/Z0TNfysTazc?si=d-b-ehf-6NQZTcCP&t=4300) [Lee Smolin - How Can Space and Time be the Same Thing? - YouTube](https://www.youtube.com/watch?v=QOAcQCFNtbo) "new political party should emerge centered around automating government, public sovereign AGI, unlocking productivity across industries/supply chains (which will enable abundance), most importantly guide the societal transition that is upon us over the next decade by revamping the operating system. things like high quality free food or housing or healthcare will no longer be out of reach when you eliminate the main bottleneck in the system, the human." https://x.com/far__el/status/1811946359014010963?t=DIoaCIfbSGUFyZoc_9SNAg&s=19 [Evolutionary game theory - Wikipedia](https://en.wikipedia.org/wiki/Evolutionary_game_theory) [Memetics - Wikipedia](https://en.wikipedia.org/wiki/Memetics) [On the mathematics of Memetics](https://dash.harvard.edu/handle/1/37372334) https://royalsocietypublishing.org/doi/10.1098/rsfs.2022.0029 boundary-cells - which fire when you are near a wall, head-direction cells that encode head orientation, object vector cells that fire when you are facing a particular direction at a particular location from a particular object etc Links for 2024-07-14 AI: 1. OpenAI is working on new reasoning technology under the code name ‘Strawberry’ (formerly known as Q*) to perform long-horizon tasks. Planning ahead will enable it to navigate the internet autonomously and reliably to perform “deep research.” In an internal all-hands meeting, OpenAI showed a demo of a research project that it claimed had new human-like reasoning skills. https://www.reuters.com/technology/artificial-intelligence/openai-working-new-reasoning-technology-under-code-name-strawberry-2024-07-12/ [archived version: https://archive.is/cnCrI] 2. "Standard deep learning (is) slow and power-hungry..we introduce..an analog electronic network (which) learns tasks unachievable in linear systems, (is) robust to damage, retrainable in seconds, & performs.. in microseconds..dissipating only picojoules" https://www.pnas.org/doi/10.1073/pnas.2319718121 3. The Making of Devin [AI software agent] by Cognition AI: Scott Wu [The Making of Devin by Cognition AI: Scott Wu - YouTube](https://www.youtube.com/watch?v=T7NWjoD_OuY) 4. “Many AI researchers believe that deep learning alone is not enough; there must be more than naive scaling to get to human level AGI. If you’re in this plurality, I have some questions” [Questions about ARC Prize | Eric Jang](https://evjang.com/2024/07/11/arc.html) 5. OpenDiLoCo, an open-source implementation and scaling of DeepMind’s Distributed Low-Communication (DiLoCo) method, enabling globally distributed AI model training. [Prime Intellect](https://www.primeintellect.ai/blog/opendiloco) 6. MambaVision: A Hybrid Mamba-Transformer Vision Backbone https://arxiv.org/abs/2407.08083 7. SEED-Story: Multimodal Long Story Generation with Large Language Model https://arxiv.org/abs/2407.08683 8. “Fast, robust, reactive, direct-from-sensor grasp-anything policies. RL really works, and it’s going to transform the entire robotics economy.” https://arxiv.org/abs/2407.02274 9. Machine Learning Can Predict Shooting Victimization Well Enough to Help Prevent It — “Out-of-sample accuracy is strikingly high: of the 500 people with the highest predicted risk, almost 13 percent are shot within 18 months, a rate 128 times higher than the average Chicagoan.” [Machine Learning Can Predict Shooting Victimization Well Enough to Help Prevent It | NBER](https://www.nber.org/papers/w30170) 10. AI system achieves 96% accuracy in determining sex from dental X-rays https://www.psypost.org/ai-system-achieves-96-accuracy-in-determining-sex-from-dental-x-rays/ 11. Helsing, a startup developing AI software for defense, raises €450 million to expand its presence in European nations bordering Russia [Bloomberg - Are you a robot?](https://www.bloomberg.com/news/articles/2024-07-11/defense-startup-helsing-nets-5-billion-valuation-plans-eastern-flank-expansion) [no paywall: https://archive.is/Ns43P] 12. Scaling Law in Neural Data: Non-Invasive Speech Decoding with 175 Hours of EEG Data. https://arxiv.org/abs/2407.07595 13. A.I. Helped Spot a Copper Bonanza. It Could Transform More Than Mining. https://www.nytimes.com/2024/07/11/climate/kobold-zambia-copper-ai-mining.html [no paywall: https://archive.is/smmDA] 14. How AI Revolutionized Protein Science, but Didn’t End It [How AI Revolutionized Protein Science, but Didn’t End It | Quanta Magazine](https://www.quantamagazine.org/how-ai-revolutionized-protein-science-but-didnt-end-it-20240626/) 15. “We introduced GPT-4, the first version, some 15 months ago. Since then, the cost of a token/ word on the model has been reduced by 85-90%. There’s no reason why that trend will not continue,” https://venturebeat.com/ai/openai-anticipates-decrease-in-ai-model-costs-amid-adoption-surge/ [archived version: https://archive.is/j0nQb] 16. GPT-2 can be trained for $672 on one 8xH100 GPU node for 24 hours. https://x.com/Jsevillamol/status/1812168535118061794 17. Neural Radiance Fields (NeRF): A full highly accurate radiance field - where you can choose to see a 3D scene from any point of view (effectively a 5D representation) - compresses the scene to roughly the size of any single training picture. [NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis (ML Research Paper Explained) - YouTube](https://www.youtube.com/watch?v=CRlN-cYFxTk) 18. “Superhuman” Go AIs still have trouble defending against these simple exploits [“Superhuman” Go AIs still have trouble defending against these simple exploits | Ars Technica](https://arstechnica.com/ai/2024/07/superhuman-go-ais-still-have-trouble-defending-against-these-simple-exploits/) AI safety: 1. Reasoning through arguments against taking AI safety seriously [Reasoning through arguments against taking AI safety seriously - Yoshua Bengio](https://yoshuabengio.org/2024/07/09/reasoning-through-arguments-against-taking-ai-safety-seriously/) 2. Towards Guaranteed Safe AI (Paper Intro) [Towards Guaranteed Safe AI (Paper Intro) - YouTube](https://www.youtube.com/watch?v=mwUfAvzjd8o) 3. On the Limitations of Compute Thresholds as a Governance Strategy https://arxiv.org/abs/2407.05694v1 4. Observational Scaling Laws and the Predictability of Language Model Performance — 1) LLM capabilities are a function of few variables 2) Using those dimensions, you can predict much better large phase change in capabilities 3) Post-training interventions effect size are also predictable https://arxiv.org/abs/2405.10938 Catastrophic and existential risks: 1. Food without agriculture: Food from CO2, biomass and hydrocarbons to secure humanity's food supply against global catastrophe https://www.sciencedirect.com/science/article/abs/pii/S0924224424002851 2. “I'm often asked about how the existential risk landscape has changed in the years since I wrote The Precipice. Earlier this year, I gave a talk on exactly that, and I want to share it here.” [The Precipice Revisited — EA Forum](https://forum.effectivealtruism.org/posts/iKLLSYHvnhgcpoBxH/the-precipice-revisited) Technology: 1. This Enormous Computer Chip Beat the World’s Top Supercomputer at Molecular Modeling [This Enormous Computer Chip Beat the World's Top Supercomputer at Molecular Modeling](https://singularityhub.com/2024/07/09/this-enormous-computer-chip-beat-the-worlds-top-supercomputer-at-molecular-modeling/) 2. A High-Energy Proton Beam Could Finally Connect Us to Proxima Centauri [High-Energy Proton Beam Could Connect Us to Proxima Centauri](https://www.popularmechanics.com/science/a61548076/proton-beam-propulsion-proxima-centauri/) 3. Animation of the path that an incoming light ray traces through the GRAVITY instrument. [Animation of the path of a light ray through GRAVITY - YouTube](https://www.youtube.com/watch?v=Zc2oMnTB08E) Politics: 1. Ice: The Penultimate Frontier — “I argue here that preventing a large iceberg from melting is absurdly cheap per unit area compared to just about any other way of making new land, and it's kind of crazy to spend money on space exploration and colonization before colonizing the oceans with floating ice-islands.” https://www.lesswrong.com/posts/gthjxPDywrMTs3p2j/ice-the-penultimate-frontier 2. Prime Minister Meloni’s government is planning to reintroduce nuclear energy 35 years after Italy shut down its last reactor. Italy plans to allow investments in small reactors as part of a bid to lower the country’s carbon emissions. [Subscribe to read](https://www.ft.com/content/a726934b-ba97-4e2c-b60d-e24a227a416f) [no paywall: https://archive.is/05ZP4] 3. "Incompetence in one figure: UN 2024 projection for India TFR: - fertility has declined at constant slope for 60 years - all recent data points toward a continuing decline - no other country succeeded at stopping TFR decline => yet UN predicts TFR to magically stabilize overnight" https://x.com/davidbessis/status/1811851769233723724 Trump assassination attempt: 1. “Given the location of the two sniper teams and the shooter, it appears that only the southern team was not completely obstructed by a tree.” https://x.com/OAlexanderDK/status/1812446084385427965 2. Read more in this GeoConfirmed with all relevant information: https://x.com/GeoConfirmed/status/1812436750066729047 Quantum mechanics is just thermodynamics in imaginary time. https://x.com/getjonwithit/status/1812573074363183494?t=4c6yqk_-F0NyAjzjrhs1Rg&s=19 " formal language has 1. TERMS (programs) 2. TYPES (specifications) 3. INFERENCE natural language has 1. WORDS (probablistic pointers to iterated abductions from qualia) 2. GRAMMAR (emergent algorithmic patterns in noospheric corpus) 3. TASTE and i think that's beautiful " https://x.com/doomslide/status/1812520040266092879?t=Eil9JSltI1bxjq0rodUGmg&s=19 LLM search is hard https://x.com/aidan_mclau/status/1812543179885027460?t=Aqv-SSrVz91DNHhzGTjlBg&s=19 [The Precipice Revisited — Toby Ord](https://www.tobyord.com/writing/the-precipice-revisited) [Discovering novel algorithms with AlphaTensor - Google DeepMind](https://deepmind.google/discover/blog/discovering-novel-algorithms-with-alphatensor/) [New HIV Prevention Drug Shows 100% Efficacy in Clinical Trial : ScienceAlert](https://www.sciencealert.com/new-hiv-prevention-drug-shows-100-efficacy-in-clinical-trial) "It’s not the populist or the mentally impaired candidate who endanger US democracy, but the monopolistic institutional process that preselects untenable, corrupted candidates as the only choices for the most important job on the planet. US democracy needs a new playbook." https://x.com/Plinz/status/1813189267314864236?t=N5NMYBrD94FjiLstA-VcRw&s=19 Blue tribe: progressives Red tribe: conservatives Gray tribe: tech libertarians Polarization chart overtime https://x.com/balajis/status/1653886507219042304?t=yLj0rI04ezEUgsBsG9l2gQ&s=19 Generalization acceleration https://arxiv.org/abs/2405.20233 Links for 2024-07-17 AI: 1. Trump allies draft AI order to launch ‘Manhattan Projects’ for defense to compete with China [Trump allies draft AI order to launch ‘Manhattan Projects’ for defense - The Washington Post](https://www.washingtonpost.com/technology/2024/07/16/trump-ai-executive-order-regulations-military) [no paywall: https://archive.is/LFseW] 2. Mistral launches Mathstral 7B and Codestral Mamba 7B. On the MATH benchmark, Mathstral 7B obtains 56.6% pass@1, outperforming Minerva 540B by more than 20%. Mathstral scores 68.4% on MATH with majority voting@64, and 74.6% using a reward model. [MathΣtral | Mistral AI | Frontier AI in your hands](https://mistral.ai/news/mathstral/) Codestral Mamba is one of the first open source models with a Mamba 2 architecture. It is the best 7B code model available, and is trained with a context length of 256k tokens. [Codestral Mamba | Mistral AI | Frontier AI in your hands](https://mistral.ai/news/codestral-mamba/) 3. SpreadsheetLLM: Encoding Spreadsheets for Large Language Models https://arxiv.org/abs/2407.09025 4. Human-like Episodic Memory for Infinite Context LLMs https://arxiv.org/abs/2407.09450 5. Generating Games Via Evolution and Language Models: “We demonstrate both quantitatively and qualitatively that our approach is capable of generating new and interesting games, including in regions of the potential rules space not covered by existing games in the Ludii dataset.” https://arxiv.org/abs/2407.09388 6. Foundational Autoraters: Taming Large Language Models for Better Automatic Evaluation https://arxiv.org/abs/2407.10817 7. Learning Multiple Concepts from a Single Image — Unsupervised Concept Extraction (UCE) is a new task that extracts and recreates multiple concepts from a single image without any human annotations. [ConceptExpress](https://haoosz.github.io/ConceptExpress/) 8. AI method radically speeds predictions of materials’ thermal properties [AI method radically speeds predictions of materials’ thermal properties | MIT News | Massachusetts Institute of Technology](https://news.mit.edu/2024/ai-method-radically-speeds-predictions-materials-thermal-properties-0716) 9. Artificial intelligence outperforms clinical tests at predicting progress of Alzheimer’s disease [Artificial intelligence outperforms clinical tests at predicting progress of Alzheimer’s disease | University of Cambridge](https://www.cam.ac.uk/research/news/artificial-intelligence-outperforms-clinical-tests-at-predicting-progress-of-alzheimers-disease) 10. Does GPT-4 have Theory of Mind? “Across the battery of theory of mind tests, we found that GPT-4 models performed at, or even sometimes above, human levels” [Testing theory of mind in large language models and humans | Nature Human Behaviour](https://www.nature.com/articles/s41562-024-01882-z) 11. Scaling Exponents Across Parameterizations and Optimizers https://arxiv.org/abs/2407.05872 Robotics: 1. “UMI on Legs is a framework for combining real-world human demonstrations with simulation trained whole-body controllers, providing a scalable approach for manipulation skills on robot dogs with arms.” [UMI on Legs: Making Manipulation Policies Mobile with a Manipulation-Centric Whole-body Controller](https://umi-on-legs.github.io/) 2. Surgical Robot Transformer🪡: Automating delicate surgical tasks with end-to-end imitation learning. [Surgical RT](https://surgical-robot-transformer.github.io/) 3. Grasping Diverse Objects with Simulated Humanoids [Grasping Diverse Objects with Simulated Humanoids](https://www.zhengyiluo.com/Omnigrasp-Site/) 4. Toward Precise Robotic Weed Flaming Using a Mobile Manipulator with a Flamethrower https://arxiv.org/abs/2407.04929 5. This $1.5 Billion AI Company Is Building A ‘General Purpose Brain’ For Robots [Skild AI Is Building A ‘General Purpose Brain’ For Robots](https://www.forbes.com/sites/rashishrivastava/2024/07/09/this-15-billion-ai-company-is-building-a-general-purpose-brain-for-robots/) [no paywall: https://archive.is/oz6ii] Biotechnology: 1. Multiplex Gene Editing: Where Are We Now? https://www.lesswrong.com/posts/oSy5vHvwSfnjmC7Tf/multiplex-gene-editing-where-are-we-now 2. Disruptive and innovative approach to drug discovery using high throughput in vivo screening. https://www.gordian.bio/blog/the-in-vivo-screening-revolution/ 3. Genomic Language Models: Opportunities and Challenges https://arxiv.org/abs/2407.11435 4. Wet-lab innovations will lead the AI revolution in biology [Wet-lab innovations will lead the AI revolution in biology](https://www.abhishaike.com/p/wet-lab-innovations-will-lead-the) Miscellaneous: 1. New quantum computer smashes 'quantum supremacy' record by a factor of 100 — and it consumes 30,000 times less power [New quantum computer smashes 'quantum supremacy' record by a factor of 100 — and it consumes 30,000 times less power | Live Science](https://www.livescience.com/technology/computing/new-quantum-computer-smashes-quantum-supremacy-record-by-a-factor-of-100-and-it-consumes-30000-times-less-power) 2. A new meta-analysis of group differences in measured IQs in Britain [The ethnic meritocracy in the United Kingdom](https://www.emilkirkegaard.com/p/the-ethnic-meritocracy-in-the-united) 3. Study reveals how an anesthesia drug induces unconsciousness [Study reveals how an anesthesia drug induces unconsciousness | MIT News | Massachusetts Institute of Technology](https://news.mit.edu/2024/study-reveals-how-anesthesia-drug-induces-unconsciousnes-0715) 4. Simulation arguments, a research paper in philosophy. https://jc.gatspress.com/pdf/simulation_arguments_revised.pdf Politics: 1. “This is from Biden’s speech to the NAACP yesterday. He loses track of what he’s saying constantly. Asking people to ignore what’s right in front of them is absurd.” https://x.com/NateSilver538/status/1813582790400872700 2. Exclusive: Secret Service ramped up security after receiving intel of Iranian plot to assassinate Trump; no known connection to shooting [Secret Service ramped up security after receiving intel of Iranian plot to assassinate Trump; no known connection to shooting | CNN Politics](https://edition.cnn.com/2024/07/16/politics/iran-plot-assassinate-trump-secret-service) 3.Thread on the toughness of various US presidents compared to Trump https://x.com/JamesSurowiecki/status/1812832482930065435 4. The Czech government has awarded a contract to build 2 new nuclear reactors (at a site of an existing NPP Dukovany) to the South Korean KHNP. The option for 2 more reactors (likely at the current NPP Temelin) will also be negotiated. https://www.reuters.com/business/energy/czechs-pick-south-koreas-khnp-over-french-bid-nuclear-power-tender-2024-07-17/ South Korea: 1. More South Koreans want Seoul to have its own nuclear weapons [Subscribe to read](https://www.ft.com/content/0a7b8855-5682-4fbf-be42-156811d4d578) [no paywall: https://archive.is/2v3Mb] 2. South Korea to mass produce lasers that can take out drones at $1.50 a hit [South Korea to mass produce lasers that can take out drones at $1.50 a hit | CNN](https://edition.cnn.com/2024/07/11/asia/south-korea-antidrone-lasers-intl-hnk-ml/index.html) Ukraine: 1. Russian state TV thinks Vance is a nightmare for Ukraine—and a dream come true for the Kremlin. [Russian state TV is giddy about JD Vance as the VP - YouTube](https://www.youtube.com/watch?v=jty49i_ZglY) 2. Ten years ago Russian separatists shot down MH17, murdering nearly 300 people, using a Buk launcher transferred from Russia on the day of the act. Russia then issued a blizzard of complete lies to conceal this fact, a charade it maintains to the present. [The truth about Putin’s shootdown | The Economist](https://www.economist.com/interactive/christmas-specials/2023/12/20/the-truth-about-putins-shootdown) [no paywall: https://archive.is/C99Nf] [SciCode - SciCode Benchmark](https://scicode-bench.github.io/) https://www.lesswrong.com/posts/JycQ8dZwAKcgfbWig/misnaming-and-other-issues-with-openai-s-human-level https://x.com/omarsar0/status/1814135222562165104?t=tzO3ajQM4T5YPMC0sEq0xg&s=19 https://arxiv.org/abs/2407.12994