DeepMind walks back AlphaFold claim, Periodic Labs hits $7B pre-shipping

Sources

Why AlphaFold Didn’t Solve Protein Folding — Pushmeet Kohli, Google DeepMind & Sal Candido, Biohub latent.space

From the Bitter Lesson of AI scaling to the unsolved mysteries of protein folding, Google DeepMind’s Pushmeet Kohli and Biohub’s Sal Candido are rethinking what it takes to build AI that truly understands biology.

Synthesis Superintelligence: from Semiconductors to Superconductors — Periodic Labs’ Liam Fedus and Ekin Dogus Cubuk latent.space

A special Science pod and Engineering pod crossover.. with Forward Deployed Engineering kicker!

References

PMC / peer-reviewed AF3 benchmark pmc.ncbi.nlm.nih.gov

AlphaFold 3 showed a 51% error rate in predicting D-peptide chirality, suggesting it struggles with molecules outside its primary training data.

PR Newswire — Virtual Biology Initiative announcement prnewswire.com

International cross-sector collaboration commits nearly $2 billion to build foundational data for AI models to predict and treat disease, anchored by DOE, NIH, CZ Biohub, Google DeepMind and Meta, with a one-year embargo on commercially generated datasets before open release.

CZ Biohub blog — rBio reasoning model biohub.org

rBio is a reasoning model trained on virtual cell simulations, letting biologists ask mutation-related questions in plain language and receive logic-based predictions about gene interactions.

BonvinLab AlphaFold teaching materials bonvinlab.org

AlphaFold typically predicts a single static snapshot and often defaults to a closed conformation even for apo proteins, obscuring binding sites relevant to drug discovery.

Hyper.ai review of AlphaFold limitations hyper.ai

AlphaFold frequently predicts the same native structure for variants known to cause misfolding in vivo and ‘hallucinates’ structured loops where flexible tethers actually exist.

Briefings in Bioinformatics (OUP) survey academic.oup.com

Because AlphaFold relies on Multiple Sequence Alignments, its accuracy drops precipitously for ‘orphan’ proteins lacking known homologs, making it closer to advanced homology modeling than a de novo physical solver.

The Stack — ‘Scant evidence: Google’s AI chemistry claims were misleading’ thestack.technology

Scant evidence for the required ‘trifecta of novelty, credibility, and utility’… DeepMind eventually clarified in a 2026 correction that ‘novel’ meant new to their internal platform, rather than undiscovered by human science.

Microsoft Research — MatterGen announcement microsoft.com

MatterGen, a generative AI model that directly designs crystalline materials based on target properties… bypassing the need to search through pre-existing databases.

StartupFortune — Alibaba DAMO ‘Elements Claw’ startupfortune.com

Elements Claw agent… screened 2.4 million structures in just 28 hours [and] led to the physical synthesis and experimental verification of four brand-new superconductors: Hf21Re25, Zr4VRe7, HfZrRe4, and Zr3ScRe8.

APNIC blog — ‘The case for open-weight models’ blog.apnic.net

FrontierXRD remains an internal, proprietary evaluation that has not yet undergone external peer review… without third-party access to the same test environment—specifically the proprietary ‘Periodic Harness’—critics argue these figures should be treated as vendor-reported claims.

Daily Dropout — Periodic Labs profile dailydropout.fyi

Valuation has reportedly surged from $1.3 billion to over $7 billion in less than a year, despite… the firm has yet to announce any significant discovery in its target fields of superconductors or semiconductors.

AI Breaking Wire — Latent Space episode recap aibreakingwire.com

Fedus provided a notable dissent from the current trend toward general-purpose humanoid robotics in science… the decade-long engineering challenge of perfecting bipedal dexterity would only delay scientific progress. Instead, Periodic Labs focuses on automating scientific instruments directly.