So much exciting news coming out of Axiom! So glad we got to sit down with you! @axiommathai @CarinaLHong

A spicy podcast. In the movie The Man Who Knew Infinity, Ramanujan became a much stronger mathematician after the Hardy made him write proofs. Verification is for better generation, for optimal design. Verified AI accelerates superintelligence, esp agentic superintelligence.

Our paper on ESMC, ESMFold2, and mechanistic interpretability for proteins is up on @biorxivpreprint! We've made a few changes since the initial version went online last week. 1. We found an issue in the way we provided MSAs to OpenFold3. This led us to report lower performance of OpenFold3 on some benchmarks. This issue does not affect any of the other models evaluated. 2. We updated how we report results on Runs N' Poses to more closely match the original paper (counting only ligands with valid SuCOS similarity score). We also add a bar plot to the supplement that stratifies performance by similarity. This mostly changes the absolute values of the pass rate, not the relative performance of models. 3. Added some more BLI data to the supplement. 4. Added some missing citations, fixed typos, etc. Check out the preprint here: biorxiv.org/content/10.64898…

event: luma.com/ls?e=evt-UV7RZqcrQa… paper: biohub.org/news/world-model-…

I love it when a plan comes together! Congrats Ron, this is a huge milestone.

"genes are code" is always vague I like: cell nucleus → storage device / storage controller ribosome → JIT-compiler and runtime features from a world model (use a SAE) → functions Proteins → processes signaling pathways → workflows Phenotypes → behaviors / outputs @biohub

The ESMC/ESMFold2 release also includes ESM Atlas, which makes me very happy :). @biohub team delivers the data!

Love this summary of the paper. The paper is definitely worth the read!

🔬Doing Vibe Physics The full story of how GPT‑5.x derived new results in theoretical physics and quantum gravity, live on our Science pod today! latent.space/p/lupsasca our conversation with @ALupsasca, an award winning theoretical physicist on his AGI-pilling journey applying GPT5 to physics problems (with a nudge from @markchen90)! Timestamps 0:00 Introduction to Al's impact on physics research 0:43 Guest introduction: Alex Luposka 2:49 Alex joining OpenAl and the shift in physics research 4:08 The release of GPT-5 and the shift in capabilities 10:05 Explaining Quantum Field Theory and amplitude calculations 14:20 Overview of gluons and the strong force 14:38 Discussing the first research paper on single-minus gluon tree amplitudes 20:56 How ChatGPT helped solve a year-long physics puzzle 23:02 Complexity of manual calculations in physics 26:12 The history and mechanics of Feynman diagrams 27:44 The Parke-Taylor formula and the quest for simplification 31:26 Using ChatGPT to find the simplification in the special phase space region 38:07 Proving the formula from scratch to ensure validity 41:00 Determining the scientific impact and future research 42:27 Introduction to the second paper on graviton amplitudes 45:41 | Defining particles, irreducible representations, and symmetry 47:46 How GPT Pro generalized the research to gravity 53:57 The epistemological shift: Is this a new way of doing physics? 59:27 The use of Al as a 'scout' for research directions 1:01:44 The role of 'taste' and collaboration with Al 1:10:23 Personal evolution from Al skeptic to resident scientist 1:12:46 Solving a black hole perturbation problem with GPT-5 1:16:34 Discussing whether Al can make original, conceptual leaps 1:20:09 Challenges of 'Al slop' and the future of academic publishing 1:23:13 The bottleneck of writing academic papers 1:30:19 Final takeaways and looking ahead to the next year

🔬 Training Transformers to solve 95% failure rate of Cancer Trials the AI for Science pod is back with @RonAlfa, CEO of @NOETIK_ai, and Daniel Bear, VP Research at Noetik, explaining exactly how their team of top AI x Bio researchers and engineers (shoutout @owl_posting) will use AI to cure cancer, by focusing on key bottlenecks like patient selection, and training large cancer foundation models like TARIO-2, an autoregressive transformer trained on one of the largest sets of tumor spatial transcriptomics datasets in the world... which first required years of blind faith in collecting good data to even get going:

@andrewwhite01/@EdisonSci- I know you just released a new literature agent. Does it read graphs? @wellingmax/@cusp_ai - I think your agents use literature too. What are you doing? 3/n

but found a big discrepancy between the plots in the reports and the interpretation. This is a good 60 seconds for anyone building scientific models based on literature. 2/n