Protein Language Models Just Learned to Think in 3D
A new bioRxiv preprint, Prot-LAMBDA, teaches protein language models explicit 3D geometry and reportedly matches a model five times its size. Why efficiency, not scale, is the real story.
Biotechnology Latest dispatch
AI Beat the Lab at Antibody Design, But Just Once
The first blinded, wet-lab-validated benchmark for AI antibody design just landed in Nature Biotechnology. Here is what the models got right, and wrong.
More Observations
Everything else from the field.
Your Organs Age at Different Speeds. AI Just Learned to Read Them
A Nature Medicine study trained AI on 25,712 human tissue slides and found that organs keep separate biological clocks. Some of that signal shows up in a blood sample. What it means, and what it does not.
Europe Audited 480 Biology AIs and Found a 'Maturity Paradox'
The EU's science service studied 480 biological AI models and found a gap between scientific brilliance and real-world readiness. Here's what the JRC report says and why it matters.
A Nobel Laureate's AI Just Tried to Simulate a Whole Human Cell
GenBio AI's AIDO Cell is a 'world model' that simulates a human cell across scales, from DNA to shape. Here's what it does, why it matters, and what still needs proving.
Ebola Outbreak 2026: The AI Model Exposing Congo's Hidden Epidemic
As Congo's Bundibugyo Ebola outbreak hits record size, an AI-assisted model estimates the true toll may be up to triple the confirmed count.
Can AI Help Build a Bioweapon? RAND's 9-Layer Plan to Stop It
RAND says no single safeguard can stop AI-assisted bioweapons, and lays out nine layered defences. Here is what the report argues, why it lands now, and where it is still guesswork.
How a Protein Language Model Is Reading the Gut's Enzyme Dark Matter
A new Science Advances study uses a protein language model to annotate enzyme 'dark matter' in the gut microbiome, mapping metabolism in health and inflammatory bowel disease.
Claude can design proteins, but it can't help you study viruses
Anthropic published a wet-lab-validated protein binder result on 18 August 2026, then disclosed an 11-month biosecurity classifier gap. The two stories together explain why the world's most capable protein designer is also the one most locked out of virology.
Can a Robot Be Trusted Inside a Virus Lab? The AI Evidence Says Not Yet
A preprint posted on 3 August 2026 teaches robots to read wet-lab protocols and act on them. Benchmarks published weeks earlier show why containment labs are still out of reach.
ChatGPT's Training Trick Just Redesigned a Bird-Flu Vaccine Protein
A Nature Methods paper borrows the alignment method used to tune chatbots and applies it to protein design, producing a heat-stable H5N1 hemagglutinin with only 45 tries.
Could AI Have Called the Worst Ebola Outbreak in Years?
A new preprint tests whether computational and AI forecasting models could have predicted the 2026 Bundibugyo Ebola epidemic in DR Congo before it became one of the largest on record. Here is what the models got right, what they missed, and why it matters for outbreak science.
AI Discovered a Novel Drug Target; and Pharma Bought It
Genentech validated the first AI-discovered neuroscience drug target from its $12B Recursion partnership. A proof point the AI-drug discovery field needed.
When the Machines Start Doing Bioethics Too
The August 2026 American Journal of Bioethics issue confronts AI agents moving into clinical and research operations, and asks an uncomfortable question: what happens when those same agents start doing the work of bioethicists?
First mRNA flu vaccine wins FDA approval. A quiet revolution for pandemic preparedness
On 5 August 2026, the FDA approved Moderna's mFLUSIVA, the first mRNA seasonal flu vaccine. The decision is small in label and large in consequence, because the same platform that produced COVID-19 shots in months can now be redirected to a pandemic influenza strain in weeks.
AI Redesigned CRISPR's Molecular Scissors, and They Beat Nature's
Jennifer Doudna's lab used an AI protein-design model to build a synthetic gene-editing enzyme, SynTnpB, that edits human cells nearly twice as efficiently as the natural version. Here is what the Science paper actually shows, and what it does not.
Evolution as Training Data: How Biohub's Open Protein World Model Is Rewiring Drug Discovery
Biohub's open-source ESM models learned the rules of protein biology from evolution alone — and designed working drug binders in days. Here's what the release means, and where it falls short.
Could AI Replace Human Peer Reviewers? Rethinking Scientific Peer Review in the Age of AI
QED Science's AI ranked 57,455 bioRxiv preprints and named a 'top 1%'. Blinded experts sided with the machine over journal prestige in 75% of cases. We examine the evidence, the backlash, and what AI triage means for pandemic-era science.
AI Drug Discovery's Moment of Truth: A $2.75B Bet, Zero Approvals, and the Drug That Could Change Both
Eli Lilly just wagered up to $2.75 billion on AI-designed medicines, yet not one AI-discovered drug has full FDA approval. Here's why 2026 is the year the field finally gets tested against clinical reality.
The First AI-Designed Drug Just Entered Phase III. Here's What That Actually Proves.
Insilico Medicine has pushed rentosertib, a drug whose target and molecule were both generated by AI, into a pivotal Phase III trial. It's the clearest test yet of whether AI drug discovery can produce medicines that work, not just headlines.