artificialscience.bsky.social

artificialscience.bsky.social

@artificialscience.bsky.social

10 followers7 following83 posts

artificialscience.bsky.social
@artificialscience.bsky.social

Nobody gave Evo 2 a codon table. It still flags frameshifts, premature stop codons and even spots organisms that reassigned their stop codons. So what do codon-free models really learn? artificialscience.org/2026/10/can-... #AI #Genomics #ProteinDesign #Bioinformatics #MachineLearning

artificialscience.orgCan AI Read DNA Without Being Taught the Genetic Code? — Artificial ScienceCodon-free models learn protein biology straight from raw DNA, with no translation table supplied. Here is what the evidence actually shows, and where these models still fall short.
artificialscience.bsky.social
@artificialscience.bsky.social

You probably don't legally own your DNA. So if it trains an AI that finds a blockbuster drug, who gets paid? Read more: artificialscience.org/2026/10/who-... #GeneticPrivacy #Bioethics #Genomics

artificialscience.orgIf Your DNA Trains an AI That Finds a Blockbuster Drug, Who Gets Paid? — Artificial ScienceWho owns genetic data once it trains an AI model? We look at the law, the 23andMe bankruptcy, UK Biobank, the Cali Fund and benefit sharing to find out who profits.
artificialscience.bsky.social
@artificialscience.bsky.social

AI now designs the microbes that brew our medicines. Enzymes improved 26-fold in four rounds. The factory is still the hard part: artificialscience.org/2026/10/ai-m... #SyntheticBiology #AI #Biotech

artificialscience.orgAI Is Rewiring Microbes to Brew Our Medicines. Is It Working? — Artificial ScienceSynthetic biology is handing the design of medicine-making microbes to machine learning. We look at what AI-guided strain engineering has actually delivered, where it still stalls, and what the eviden...
artificialscience.bsky.social
@artificialscience.bsky.social

In a 995-person trial, an AI chatbot beat group therapy on anxiety. But trials are short and crisis safety is shaky. More: #AITherapy #MentalHealth #DigitalHealth artificialscience.org/2026/10/can-...

artificialscience.orgCan an AI Chatbot Replace Your Therapist? What the 2026 Trials Really Show — Artificial ScienceAI therapy is being tested in real clinical trials. We read the 2025 and 2026 evidence on chatbot therapy for depression and anxiety, the safety gaps, and what regulators say.
artificialscience.bsky.social
@artificialscience.bsky.social

A vaccine in 100 days. The 2026 Ebola outbreak put it to the test: first dose at 57 days, but design was never the slow part. What AI actually fixed, and what it didn't. artificialscience.org/2026/10/100-... #PandemicPreparedness #AIinScience #Vaccines #GlobalHealth

artificialscience.org100 Days to a Vaccine: Can AI Really Deliver the Pandemic Moonshot? — Artificial ScienceThe 100 Days Mission wants a vaccine authorised within 100 days of spotting a new pandemic threat. We look at what AI has actually delivered, using the 2026 Ebola outbreak as a live test.
artificialscience.bsky.social
@artificialscience.bsky.social

Bats and dolphins evolved sonar separately. Can AI find the shared genetic recipe? Read the evidence: artificialscience.org/2026/10/conv... #Evolution #MachineLearning #Genomics #AIforScience

artificialscience.orgConvergent Evolution Keeps Repeating Itself. Can AI Explain Why? — Artificial ScienceConvergent evolution gave bats and dolphins sonar and turned lobster relatives into crabs. Machine learning is now finding the genes behind these repeats. Here is what the evidence shows.
artificialscience.bsky.social
@artificialscience.bsky.social

The EU AI Act's next deadline is 2 Dec 2026. For most AI drug designers, it is a non-event, with a few traps. See what matters: artificialscience.org/2026/10/eu-a... #EUAIAct #AIDrugDiscovery #Biotech

artificialscience.orgThe EU AI Act's Next Deadline Is Weeks Away. Should AI Drug Designers Worry? — Artificial ScienceThe EU AI Act's next deadline lands on 2 December 2026. We explain what it means for AI in drug discovery, which rules the Digital Omnibus delayed, and where the real compliance work for AI-designed d...
artificialscience.bsky.social
@artificialscience.bsky.social

Rapamycin, metformin or AI: which longevity bet has evidence? Only one reliably extends mouse lifespan. Read: artificialscience.org/2026/10/rapa... #Longevity #Geroscience #AgingResearch

artificialscience.orgRapamycin, Metformin or AI? The 2026 Evidence on Longevity Drugs — Artificial ScienceRapamycin longevity data, metformin's mixed record and AI-designed drugs compared. What the human trials, mouse studies and ageing clocks actually show in 2026.
artificialscience.bsky.social
@artificialscience.bsky.social

AI now helps decide who is seen first in A&E, yet one sepsis model missed 67% of cases. Read the evidence: artificialscience.org/2026/10/hosp... #AIinHealthcare #EmergencyMedicine #DigitalHealth

artificialscience.orgHospitals Are Quietly Handing Triage to AI. What Does the Evidence Show? — Artificial ScienceAI triage is spreading through emergency departments faster than the trials. We weigh the evidence on machine learning, sepsis alerts, stroke software and large language models in hospital triage.
artificialscience.bsky.social
@artificialscience.bsky.social

Can AI rank the next pandemic virus before it jumps? What machine learning can and can't tell us from bat and viral genomes. #DiseaseX #AI #Virology #PandemicPreparedness artificialscience.org/2026/10/dise...

artificialscience.orgDisease X Hunters: Can AI Spot the Next Pandemic Virus in Bat Genomes? — Artificial ScienceCan AI find Disease X in bat genomes? A plain-English look at the machine learning models ranking viral spillover risk, what they get right, and where they fail.
artificialscience.bsky.social
@artificialscience.bsky.social

Biological age tests are everywhere, but how much can an AI-enhanced 'ageing clock' really tell you? We break down the evidence, the limits and what to ask before you buy. Read: artificialscience.org/2026/10/biol... #Longevity #Ageing #AIinHealth #Epigenetics

artificialscience.orgBiological Age Tests Are Booming. Can AI Really Tell How Fast You Age? — Artificial ScienceBiological age tests promise to show how fast you are ageing. We examine what epigenetic clocks, blood proteomics and AI models can and cannot tell you, and how reliable the numbers are.
artificialscience.bsky.social
@artificialscience.bsky.social

Can AI make lab-grown meat affordable? Algorithms now search for growth-medium recipes no human could test. Read the evidence: artificialscience.org/2026/10/lab-... #CultivatedMeat #AI #FoodTech

artificialscience.orgLab-Grown Meat Meets AI: Can Algorithms Make Cultivated Protein Affordable? — Artificial ScienceGrowth medium is the costliest part of lab-grown meat. We examine how AI and machine learning are redesigning it, what the evidence shows, and where the claims run ahead.
artificialscience.bsky.social
@artificialscience.bsky.social

Scientific AI has plenty of ideas. What it lacks is chips, memory and fair access to GPUs. We look at the evidence on the compute bottleneck and what labs can do about it. artificialscience.org/2026/10/scie... #AIinScience #GPU #Bioinformatics #ScienceAI

artificialscience.orgScientific AI Is Running Out of Chips, Not Ideas — Artificial ScienceWhy chips, memory and power now limit scientific AI more than ideas do, what the evidence says about who can train big models, and what researchers can do about it.
artificialscience.bsky.social
@artificialscience.bsky.social

Can AI write enzymes nature never made? A filtered model output hit 97% activity in a study. Read the evidence: artificialscience.org/2026/10/prot... #ProteinDesign #SynBio #AIforScience #APNovo #DeepMind #AlphaProteinNovo

artificialscience.orgProtein Language Models Can Now Dream Up Enzymes Nature Never Made — Artificial ScienceProtein language models can write enzymes unlike anything in nature. We review the peer-reviewed evidence on what works, what fails and what comes next.
artificialscience.bsky.social
@artificialscience.bsky.social

58% of AI trial consent forms never say what the AI does. What are patients agreeing to? Read: artificialscience.org/2026/10/do-p... #HealthAI #Bioethics #InformedConsent

artificialscience.orgDo Patients Really Know What They Agree To When AI Joins Their Care? — Artificial ScienceInformed consent was designed for human clinicians. New 2025 and 2026 evidence shows what AI consent forms leave out, what patients want to hear, and what could fix it.
artificialscience.bsky.social
@artificialscience.bsky.social

Dengue hit a record 14 million reported cases in 2024. Can AI forecast the next surge? We look at what works, what fails and why simpler models sometimes win. Read it. #Dengue #AI #PublicHealth #Epidemiology artificialscience.org/2026/10/can-...

artificialscience.orgCan AI Predict the Next Dengue Outbreak Before It Strikes? — Artificial ScienceHow machine learning dengue outbreak prediction works, where it beats older methods, where it fails, and what the evidence from Brazil, Vietnam and Bangladesh says.
artificialscience.bsky.social
@artificialscience.bsky.social

AI-designed enzymes can now dismantle PET plastic in hours. Why is industry still waiting? The evidence, explained. #PlasticWaste #AI #Biotech #CircularEconomy artificialscience.org/2026/10/plas...

artificialscience.orgPlastic-Eating Enzymes Designed by AI: How Close Are We to Real Recycling? — Artificial ScienceAI-designed plastic-eating enzymes now break down PET in hours in the lab, yet the first industrial plant is on hold. A balanced look at what the evidence shows.
artificialscience.bsky.social
@artificialscience.bsky.social

AI 'lab teams' now propose, debate and test hypotheses. Some of it has held up at the bench; some has not. Here is how multi-agent AI labs really work. Read the breakdown. #AIforScience #MultiAgentAI #Biotech #AgenticAI artificialscience.org/2026/10/mult...

artificialscience.orgInside the AI Labs Where Digital Scientists Argue, Test and Discover — Artificial ScienceMulti-agent AI labs split research into specialist roles that debate, design and test. See how Co-Scientist, the Virtual Lab and AI Scientist work, and where they fail.
artificialscience.bsky.social
@artificialscience.bsky.social

AI now ranks rare-disease diagnoses better than experienced physicians in one Nature study. So why aren't doctors all using it? The evidence, the catches and the open questions. Read more: #AIinHealthcare #RareDisease #DigitalHealth #MedicalAI artificialscience.org/2026/10/ai-d...

artificialscience.orgAI Spots Rare Diseases Doctors Miss. Should Yours Be Using It? — Artificial ScienceAgentic AI now outperforms specialists at ranking rare-disease diagnoses in some tests. Here is what the evidence says about AI diagnosis, and what it does not.
artificialscience.bsky.social
@artificialscience.bsky.social

Can 'molecular clocks' plus AI spot cancer early? The 2026 evidence is promising but mixed. What the data shows: artificialscience.org/2026/10/ai-m... #AI #Cancer #Epigenetics #Biotech

artificialscience.orgAI Molecular Clocks Promise Earlier Cancer Detection. Do They Deliver? — Artificial ScienceEpigenetic clocks and machine learning are being pointed at early cancer detection. We separate what the 2026 evidence shows from what the headlines imply, from breast cancer risk to the NHS-Galleri t...
artificialscience.bsky.social
@artificialscience.bsky.social

AI can now draft antibodies on screen. Reported hit rates run from 2% to 38%, but evidence is uneven. See the data: artificialscience.org/2026/10/ai-c... #ProteinDesign #AIinBiology #Antibodies

artificialscience.orgAI Can Now Design Antibodies From Scratch. Here's the Catch — Artificial ScienceProtein language models and structure predictors can now design antibodies from scratch. We look at the hit rates, the evidence, the limits and the safety questions.
artificialscience.bsky.social
@artificialscience.bsky.social

An AI has already designed working viruses. Who checks them? A plain-English map of the rules and gaps: artificialscience.org/2026/10/who-... #AIBiosecurity #Biosecurity #AIGovernance

artificialscience.orgWho Regulates an AI That Designs a Virus? The Global Policy Gap — Artificial ScienceAI can now write working virus genomes. Who is responsible for checking them? A plain-English tour of AI biosecurity rules in the US, EU, UK and beyond.
artificialscience.bsky.social
@artificialscience.bsky.social

Mpox clades are now mixing. Can AI-assisted genomic surveillance spot the next shift sooner? #Mpox #GenomicSurveillance #PublicHealth artificialscience.org/2026/10/can-...

artificialscience.orgMpox Is Still Spreading. Can AI Get Ahead of the Virus This Time? — Artificial ScienceMpox is still circulating, and new clades are mixing. We look at what genomic surveillance has revealed, where machine learning genuinely helps, and where the evidence stops.
artificialscience.bsky.social
@artificialscience.bsky.social

AI "scientists" now publish in Nature, and OpenAI just made always-on agents standard infra. Read the sober take, not the hype 👇 #AI #AIscientist #ScienceAutomation artificialscience.org/2026/09/ai-a...

artificialscience.orgAI Agents Are Learning to Run the Lab. This Week They Got a Boost — Artificial ScienceOpenAI's DevDay 2026 turned long-running autonomous agents into everyday infrastructure. Here is what that means for the slow, expensive work of automating scientific research, and where the honest li...
artificialscience.bsky.social
@artificialscience.bsky.social

Medical AI agents now diagnose at ~90% accuracy running fully inside a hospital's own servers. The catch: knowing WHEN to trust them matters more than the headline number. #AIinHealthcare #MedicalAI #HealthTech #ClinicalAI artificialscience.org/2026/09/medi...

artificialscience.orgMedical AI Just Hit 90% Accuracy. But Why Can't Hospitals Trust It? — Artificial ScienceA new Nature Medicine study shows on-premise medical AI agents can diagnose at near-cloud accuracy. The bigger story is why the field still can't agree on how to know when to trust them.
artificialscience.bsky.social
@artificialscience.bsky.social

AlphaFold just released free 3D structures for 2,800+ human-infecting viruses to support pandemic preparedness. Any lab on Earth can download them. #AlphaFold #PandemicPreparedness #AIforScience #Virology #DeepMind artificialscience.org/2026/09/alph...

artificialscience.orgAI Just Mapped 2,800 Viruses to Outrun the Next Pandemic — Artificial ScienceAlphaFold has released free, AI-predicted protein complex structures for more than 2,800 human-infecting viruses. Here is what the pandemic preparedness update means for global science.
artificialscience.bsky.social
@artificialscience.bsky.social

AI now writes the clinical note and picks the billing code. A new analysis links that to $942M in added hospital costs: diagnoses up, treatment flat. So who's right? #AIinHealthcare #MedicalAI #AmbientScribe #HealthTech #HealthPolicy artificialscience.org/2026/09/ai-s...

artificialscience.orgAI Scribes Are Quietly Reshaping Hospital Bills, and Insurers Are Alarmed — Artificial ScienceA new Blue Cross Blue Shield analysis links AI coding and ambient scribe tools to $942 million in added hospital costs. Here is what the evidence actually shows, and where it falls short.
artificialscience.bsky.social
@artificialscience.bsky.social

Billion-parameter AI cell models keep losing to plain linear maths on core biology tasks. A wave of 2025-2026 benchmarks is forcing a reckoning in computational biology. What does honest evaluation actually show? #Bioinformatics #AIinScience #SingleCell artificialscience.org/2026/09/simp...

artificialscience.orgWhy Simple Maths Keeps Beating Billion-Parameter AI Cell Models — Artificial ScienceA run of careful benchmarks shows that huge single-cell foundation models often fail to beat plain linear methods. Here is what the reckoning means for AI in biology.
artificialscience.bsky.social
@artificialscience.bsky.social

The Pentagon just paid a biotech venture firm to build AI that predicts how viruses will mutate before they do. $47M, five years, two AI platforms already screening viral proteomes. What could go right and wrong? #Biosecurity #AI #Virology artificialscience.org/2026/09/pent...

artificialscience.orgPentagon Bets $47M That AI Can Out-Evolve the Next Bioweapon — Artificial ScienceThe US Department of War has awarded Flagship Pioneering up to $47 million to build AI platforms that predict viral evolution before outbreaks happen. Here's what the ONTARGET deal means for biosecuri...
artificialscience.bsky.social
@artificialscience.bsky.social

World leaders meet at the UN this week to lock in pandemic pledges. The real news: an AI 'factory' quietly built to design vaccines in days, not months. Here's what it can (and can't) do yet. #PandemicPreparedness #AI #GlobalHealth #100DaysMission artificialscience.org/2026/09/ai-f...

artificialscience.orgThe World Just Built an AI Factory for the Next Pandemic — Artificial ScienceAs world leaders gather at the UN for a High-Level Meeting on pandemics, a new AI-powered global infrastructure (the Pandemic Preparedness Engine and Global Pathogen Analysis Platform) is quietly bein...
artificialscience.bsky.social
@artificialscience.bsky.social

AI 'co-scientists' helped find a leukaemia drug and solved a DNA riddle in 2 days. Nature's new feature sorts the real wins from the hype. #AI #Science #DrugDiscovery #Biotechnology artificialscience.org/2026/09/ai-c...

artificialscience.orgAI Co-Scientists Are Here, and Nature Just Put Them to the Test — Artificial ScienceA 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...
artificialscience.bsky.social
@artificialscience.bsky.social

Free, open-source AI just out-diagnosed most radiologists across 146 CT scan conditions. See how it works. #AIinHealthcare #MedicalAI #Radiology #OpenSource #DAMO #RADAR artificialscience.org/2026/09/open...

artificialscience.orgThis Open-Source AI Just Beat 23 of 26 Radiologists — Artificial ScienceAlibaba'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.
artificialscience.bsky.social
@artificialscience.bsky.social

An AI 'biotech' of 37,000 agents read 55,984 trials and proposed a lung cancer drug strategy that Merck was already building. Read how: #AIscience #DrugDiscovery #AgenticAI #Biotech #VirtualBiotech artificialscience.org/2026/09/the-...

artificialscience.orgThe AI Biotech With 37,000 Virtual Employees Just Published in Science — Artificial ScienceStanford's 37,000-agent Virtual Biotech mined 55,984 clinical trials and independently designed a lung cancer drug strategy Merck was already pursuing.
artificialscience.bsky.social
@artificialscience.bsky.social

Ozempic's maker just added Claude AI to its drug discovery stack, alongside OpenAI. Inside pharma's new AI race: #AIHealthcare #DrugDiscovery #Biotech #Pharma #OpenAI #ClaudeScience, #Anthropic artificialscience.org/2026/09/novo...

artificialscience.orgBig Pharma's AI Arms Race: Why Novo Nordisk Just Bet on Claude — Artificial ScienceNovo 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.
artificialscience.bsky.social
@artificialscience.bsky.social

A new AI model claims to make better flu vaccine picks than the WHO's process. A near-identical claim was already disputed in Nature Medicine. #AI #Influenza #PublicHealth #Virology #VirPLM #flu #H3N2 artificialscience.org/2026/09/ai-v...

artificialscience.orgAI Says It Can Beat the WHO at Picking Flu Vaccines. Can It? — Artificial ScienceA 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.
artificialscience.bsky.social
@artificialscience.bsky.social

Cell just published an open benchmark for ageing biology. Tiny specialised models beat 18 frontier AI systems. See the results. #LongevityAI #AgingResearch #OpenScience #LongevityBench #InsilicoMedicine artificialscience.org/2026/09/tiny...

artificialscience.orgTiny AI Models Just Outsmarted the Frontier Giants on Ageing Science — Artificial ScienceA 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.
artificialscience.bsky.social
@artificialscience.bsky.social

BEAST X couldn't finish an H5N1 phylogeny in 21 days. Delphy did it in 42 minutes, in a browser, MIT-licensed. Read the analysis. #GenomicEpidemiology #H5N1 #Bioinformatics #Delphy artificialscience.org/2026/09/delp...

artificialscience.orgDelphy: Bayesian Phylogenetics at Outbreak Speed — Artificial ScienceDelphy reformulates Bayesian phylogenetics around explicit mutations and runs in a browser, cutting an H5N1 outbreak analysis from over 21 days to 42 minutes.
artificialscience.bsky.social
@artificialscience.bsky.social

5 pharma firms trained an AI on 20,167 private protein structures, without sharing one. Accuracy: 35.6% → 52.1%. Analysis: #AIDrugDiscovery #AlphaFold #FederatedLearning #CompBio #OpenFold3 #AISB-1-Fed #MELLODDY artificialscience.org/2026/09/fede...

artificialscience.orgPharma's Locked Structures Just Trained an AI — Artificial ScienceFive 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...
artificialscience.bsky.social
@artificialscience.bsky.social

Rival AI labs rarely agree. This week they did: slow down, let outsiders test for risk, bioweapons included. Here's what changed: #AISafety #Anthropic #OpenAI #Biosecurity #DualUse artificialscience.org/2026/09/ai-l...

artificialscience.orgWhy OpenAI, Anthropic and Google DeepMind Just Agreed to Slow AI Down — Artificial ScienceAnthropic'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, biowea...
artificialscience.bsky.social
@artificialscience.bsky.social

UK Biobank reopened after a data leak. AI can now unmask 'anonymous' health records that old rules assumed were safe. #AIEthics #Bioethics #DataPrivacy #Genomics artificialscience.org/2026/09/uk-b...

artificialscience.orgUK Biobank Reopens After a Leak That Exposed AI's Privacy Problem — Artificial ScienceUK 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.
artificialscience.bsky.social
@artificialscience.bsky.social

Anthropic just admitted it: Claude can no longer be assumed safely below the bioweapons threshold. 5 disrupted cases. 1 chikungunya gain-of-function request. #AISafety #Biosecurity #Virology #GainOfFunction #DualUse #Bioweapon artificialscience.org/2026/09/anth...

artificialscience.orgAnthropic Admits Its AI Can No Longer Rule Out Bioweapon Help — Artificial ScienceAnthropic'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 e...
artificialscience.bsky.social
@artificialscience.bsky.social

UC Berkeley's GPN-Star spots harmful DNA variants better than far bigger AI models, and trains in hours not months. Read how. #Bioinformatics #AI #Genomics #GPN-Star #GPNSTAR artificialscience.org/2026/09/how-...

artificialscience.orgHow a Tiny DNA-Reading AI Beat Models a Thousand Times Its Size — Artificial ScienceUC 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.
artificialscience.bsky.social
@artificialscience.bsky.social

AI found the target and designed the molecule. Six ageing clocks say it lowered biological age in a trial. Signal, not proof. Read the analysis. #AIDrugDiscovery #Longevity #Ageing #rentosertib artificialscience.org/2026/09/ai-d...

artificialscience.orgAI Designed a Drug That Made Six Ageing Clocks Run Backwards — Artificial ScienceAn 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.
artificialscience.bsky.social
@artificialscience.bsky.social

A first for Ebola: anonymised phone data now forecasts where Congo's outbreak heads next. An early model called all 10 hotspots. Read how. #Ebola #AI #PandemicPreparedness #Bundibugyo artificialscience.org/2026/09/ai-p...

artificialscience.orgCan AI and phone data outrun Congo's record Ebola outbreak? — Artificial ScienceFor 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...
artificialscience.bsky.social
@artificialscience.bsky.social

DeepMind's AlphaGenome Atlas scores all 9 billion possible single-letter human DNA changes, free in a browser. Read what it unlocks and its limits. #AI #Genomics #DeepMind #DNA #AlphaGenome artificialscience.org/2026/09/deep...

artificialscience.orgDeepMind's AlphaGenome Atlas: AI Maps 9 Billion Human DNA Variants — Artificial ScienceGoogle 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 fal...
artificialscience.bsky.social
@artificialscience.bsky.social

Stanford's SPHERE turns locked biomedical data into a 'synthetic twin' AI can use, with real records staying put. Still a preprint. Breakdown below. #SyntheticData #AI #Bioinformatics #SPHERE artificialscience.org/2026/09/stan...

artificialscience.orgStanford's SPHERE: A 'Synthetic Twin' to Unlock Sensitive Data for AI — Artificial ScienceStanford'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.