CRITICAL//Consensus

😎

Links for 2026-04-14 AI 1. The policy surrounding Mythos marks an irreversible power shift https://www.lesswrong.com/posts/3MhJELzwpbR42xsJ3/the-policy-surrounding-mythos-marks-an-irreversible-power 2. Before he wrote AI 2027, he predicted the world in 2026. How did he do? https://asteriskmag.substack.com/p/before-he-wrote-ai-2027-he-predicted 3. Does an imperfect verifier break reinforcement learning with verifiable rewards (RLVR)? Turns out it doesn’t!”https://arxiv.org/abs/2604.07666 4. Evidence on white-collar work displacement is beginning to match the theory. https://www.ft.com/content/b69f8599-eaf1-477a-a5a8-60a715e56a04?syn-25a6b1a6=1 [No paywall: https://archive.is/R99j2] 5. LLM-as-a-Verifier: A General-Purpose Verification Framework https://llm-as-a-verifier.notion.site/ 6. Can small quantum computers accelerate AI on massive classical data? Yes! https://arxiv.org/abs/2604.07639 7. QED-Nano: Teaching a Tiny Model to Prove Hard Theorems https://arxiv.org/abs/2604.04898v1 8. WRAP++: Web discoveRy Amplified Pretraining https://arxiv.org/abs/2604.06829 9. Scaling Coding Agents via Atomic Skills https://arxiv.org/abs/2604.05013 10. The Art of Building Verifiers for Computer Use Agents https://arxiv.org/abs/2604.06240 11. Squeeze Evolve: A Unified Framework for Verifier-Free Evolution https://squeeze-evolve.github.io/#blog-squeeze-evolve 12. Memento: Teaching LLMs to Manage Their Own Context https://microsoft.github.io/memento/blogpost/ 13. Neural Computers: Unlike conventional computers, which execute explicit programs, agents, which act over external execution environments, and world models, which learn environment dynamics, NCs aim to make the model itself the running computer. https://arxiv.org/abs/2604.06425 14. Efficient RL Training for LLMs with Experience Replay https://arxiv.org/abs/2604.08706 15. A new agentic application that makes lab-in-the-loop drug discovery accessible to every researcher https://aws.amazon.com/de/blogs/industries/introducing-amazon-bio-discovery/ 16. Gemini Robotics-ER 1.6: Powering real-world robotics tasks through enhanced embodied reasoning https://deepmind.google/blog/gemini-robotics-er-1-6/ 17. Intel, Google Deepen Collaboration to Advance AI Infrastructure https://newsroom.intel.com/data-center/intel-google-deepen-collaboration-to-advance-ai-infrastructure 18. OpenAI’s Chief Scientist on Continual Learning Hype, RL Beyond Code, & Future Alignment Directions https://www.youtube.com/watch?v=vK1qEF3a3WM 19. Elorian is building the foundation of visual reasoning https://www.youtube.com/watch?v=YlvfNpOMeOY 20. Slightly-Super Persuasion Will Do https://www.lesswrong.com/posts/JcavsPku6RR9hcujz/slightly-super-persuasion-will-do 21. Amazon CEO Jassy defends $200 billion AI spend: “We’re not going to be conservative” https://www.cnbc.com/2026/04/09/amazon-ceo-andy-jassy-ai-spending.html Anthropic 1. “If Mythos actually made Anthropic employees 4x more productive, I would radically shorten my timelines” https://www.lesswrong.com/posts/Jga7PHMzfZf4fbdyo/if-mythos-actually-made-anthropic-employees-4x-more 2. Claude Mythos #3: Capabilities and Additions https://www.lesswrong.com/posts/2ziYGFK7QmbbLgBoP/claude-mythos-3-capabilities-and-additions 3. Claude Mythos: China Reacts https://www.chinatalk.media/p/chinese-reactions-to-claude-mythos 4. Anthropic repeatedly accidentally trained against the CoT https://www.lesswrong.com/posts/K8FxfK9GmJfiAhgcT/anthropic-repeatedly-accidentally-trained-against-the-cot Technology 1. An interactive visualization of technological history from 3 million years ago to today. https://www.historicaltechtree.com/ 2. Chip Can Project Video the Size of a Grain of Sand — MEMS array to steer lasers for quantum computer finds other uses https://spectrum.ieee.org/mems-photonics 3. Scientists Grow Electronics Inside the Brains of Living Mice https://singularityhub.com/2026/04/13/scientists-grow-electronics-inside-the-brains-of-living-mice/

🔗

🔗

🔗

cool website lmk if you find anything that resonates w you https://camelot.wiki/citadel/camelot.wiki/divination/Xenoactualization

Links for 2026-04-06 AI 1. “The man betting everything on superintelligence is telling the world that this thing is coming so fast, and so hard, that capitalism as we know it won't be enough.” https://www.axios.com/2026/04/06/behind-the-curtain-sams-superintelligence-new-deal [no paywall: https://archive.is/ueOsp] 2. Anthropic’s Pause is the Most Expensive Alarm in Corporate History [Fiction] https://www.lesswrong.com/posts/d8bZFuYba4KPtzzRY/anthropic-s-pause-is-the-most-expensive-alarm-in-corporate 3. “In my experience, most people (including today’s straight-liners) start out suspicious of “straight lines on graphs,” but eventually have the intuition beaten into them.” https://joelbkr.substack.com/p/straight-lines-on-graphs 4. Announcing 1-bit Bonsai: The First Commercially Viable 1-bit LLMs https://prismml.com/news/bonsai-8b 5. Introducing Atlas RF Studio: Toward a Foundation Model for Electromagnetics https://www.arenaphysica.com/publications/rf-studio 6. MemFactory: Unified Inference & Training Framework for Agent Memory https://arxiv.org/abs/2603.29493 7. Gemma 4: Google’s new family of open models you can run on your own hardware. https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/ 8. Sam Altman’s prediction has come through https://marginalrevolution.com/marginalrevolution/2026/04/sam-altmans-prediction-has-come-through.html 9. Mustafa Suleyman says renegotiating Microsoft’s contract with OpenAI “unlocked [Microsoft’s] ability to pursue superintelligence.” https://www.theverge.com/report/905791/mustafa-suleyman-microsoft-ai-transcription-model 10. The quest for general intelligence is hitting a wall https://www.lesswrong.com/posts/xZsuBaQFGEb743RiM/the-quest-for-general-intelligence-is-hitting-a-wall 11. Advice for PhD students in economics about using AI. [PDF] https://economics.mit.edu/sites/default/files/2026-04/IA%20AI%20note_1.pdf 12. Q1 2026 Timelines Update https://www.lesswrong.com/posts/XLLjqMxETva3ABtsK/q1-2026-timelines-update Math 1. A Lean 4 formalization of An Introduction to Algebraic Combinatorics by Darij Grinberg, built on Mathlib. https://faabian.github.io/algebraic-combinatorics/targets.html 2. Free math book. “Combinatorics Through Guided Discovery,” by Ken Bogart. https://bogart.openmathbooks.org/ 3. Free math book. Elementary Number Theory: Primes, Congruences, and Secrets https://wstein.org/ent/ Neuro(tech) 1. A biologically plausible stochastic recurrent neural network that performs Bayesian inference, predictive-coding, continual learning, spontaneous memory replay, and progressively orthogonalizes its own representations? Emerging from first principles? https://www.sciencedirect.com/science/article/pii/S0925231226008696 2. The Practical Guide to Superbabies https://www.lesswrong.com/posts/PPLHfFhNWMuWCnaTt/the-practical-guide-to-superbabies-3 3. An eye prosthesis 20 years in the making restores sight in patients with a common age-related eye disease. https://stanfordmag.org/contents/vision-quest 4. A single dose of a new cancer drug made a brain tumor almost disappear – in just five days. [PDF] https://www.nejm.org/doi/pdf/10.1056/NEJMoa2314390 4. A shortage of synapses in schizophrenia? https://www.mpg.de/26333328/blood-test-schizophrenia 5. When and Why Did Human Brains Decrease in Size? A New Change-Point Analysis and Insights From Brain Evolution in Ants https://www.frontiersin.org/journals/ecology-and-evolution/articles/10.3389/fevo.2021.742639/full Miscellaneous 1. New holographic data storage approach packs more data into the same space https://www.optica.org/about/newsroom/news_releases/2026/new_holographic_data_storage_approach_packs_more_data_into_the_same_space/ 2. The upper bounds of humanity’s potential future https://www.lesswrong.com/posts/GLD5AiiQJqFbKX9vo/nick-bostrom-how-big-is-the-cosmic-endowment 3. More, and More Extensive, Supply Chain Attacks https://www.jefftk.com/p/more-and-more-extensive-supply-chain-attacks

🔗

🔗