This Open-Source AI Just Beat 23 of 26 Radiologists
Alibaba's DAMO Academy has open-sourced RADAR, a Science-published AI that outperformed 23 of 26 radiologists reading abdominal CT scans for 146 conditions, free for any hospital to run.
AI Latest dispatch
AI Co-Scientists Are Here, and Nature Just Put Them to the Test
A Nature feature published this week tracks how AI co-scientist platforms from Google DeepMind, FutureHouse and others are already generating validated hypotheses in real labs, and why human judgement is becoming the scarcest resource in science.
More Observations
Everything else from the field.
The AI Biotech With 37,000 Virtual Employees Just Published in Science
Stanford's 37,000-agent Virtual Biotech mined 55,984 clinical trials and independently designed a lung cancer drug strategy Merck was already pursuing.
Big Pharma's AI Arms Race: Why Novo Nordisk Just Bet on Claude
Novo Nordisk just added Anthropic's Claude Science to an AI drug discovery stack that already includes OpenAI. Here's what the deal reveals about AI in healthcare.
AI Says It Can Beat the WHO at Picking Flu Vaccines. Can It?
A new protein-language-model study claims to out-predict the WHO's flu vaccine picks, reviving a live dispute over whether AI truly beats expert judgement here.
Tiny AI Models Just Outsmarted the Frontier Giants on Ageing Science
A new Cell study pits 18 leading AI systems against ageing biology data. Small, specialised models beat every frontier system, and the whole toolkit is free to use.
Delphy: Bayesian Phylogenetics at Outbreak Speed
Delphy reformulates Bayesian phylogenetics around explicit mutations and runs in a browser, cutting an H5N1 outbreak analysis from over 21 days to 42 minutes.
Pharma's Locked Structures Just Trained an AI
Five pharmaceutical companies fine-tuned OpenFold3 on 20,167 private protein–ligand structures without pooling the data. Protein–ligand accuracy jumped from 35.6% to 52.1%. Here is what that means for drug discovery, antiviral design and pandemic preparedness.
Why OpenAI, Anthropic and Google DeepMind Just Agreed to Slow AI Down
Anthropic's Dario Amodei asked rivals to slow AI development and let outside evaluators inside frontier labs. OpenAI and Google DeepMind agreed within days. Here's what's actually being tested, bioweapons risk included.
UK Biobank Reopens After a Leak That Exposed AI's Privacy Problem
UK Biobank's Research Analysis Platform reopens this September after a data security incident, and a 2026 study shows why AI is quietly eroding old promises of genetic anonymity.
Anthropic Admits Its AI Can No Longer Rule Out Bioweapon Help
Anthropic's September 2026 threat report reveals Claude was used in five bioweapons-adjacent research attempts, including a chikungunya gain-of-function request. Here's why virologists and AI safety experts are both worried.
How a Tiny DNA-Reading AI Beat Models a Thousand Times Its Size
UC Berkeley's GPN-Star, a genomic language model published in Nature, predicts harmful DNA variants better than far larger AI models, and trains in hours not months.
AI Designed a Drug That Made Six Ageing Clocks Run Backwards
An AI-discovered, AI-designed drug lowered predicted biological age across six independent ageing clocks in a phase 2a trial. What it means, and what it doesn't.
Can AI and phone data outrun Congo's record Ebola outbreak?
For the first time in an Ebola response, WHO and Flowminder are using anonymised mobile-phone data and modelling to predict where Congo's Bundibugyo outbreak will spread next. An early test called all ten hotspots. Here is what it means for pandemic preparedness.
DeepMind's AlphaGenome Atlas: AI Maps 9 Billion Human DNA Variants
Google DeepMind's AlphaGenome Atlas precomputes the molecular effects of all 9 billion single-letter human DNA changes and offers them free in a browser. What it does, why it matters, and where it falls short.
Stanford's SPHERE: A 'Synthetic Twin' to Unlock Sensitive Data for AI
Stanford's SPHERE turns locked biomedical datasets into shareable 'synthetic twins' for AI, and opens an Alzheimer's cohort to prove it. The promise, and the caveats.
OpenEvidence's Darwin: A Medical AI Scored 100%, Then Got Locked Away
OpenEvidence's Darwin is the first medical AI to score 100% on MedQA, yet the company is withholding it over biosecurity risk. The claim, and its caveats.
The End of the Ageing Clock? How AI Learned to Reason About Getting Old
AI in 2026 is moving past ageing clocks that spit out a single number towards foundation models that reason about biological age, predict disease and nominate targets. Here is what changed and why it matters.
AI Protein Design Cracks Biotech's Biggest Bottleneck: Delivery
German scientists used generative AI to design Synthetic Transfer Vehicles, protein shells that carried RNA into cells more efficiently than the lipid nanoparticles they tested and delivered a CRISPR edit in a pig.
How a New AI Out-Predicts AlphaFold 3 on Protein Metal Ions
BiteNetI, a new deep-learning model, out-predicts AlphaFold 3 at finding protein metal-ion binding sites, a blind spot for today's protein language models.