Section
Science & Research
Toggle SidebarIAIThe Impact of AIEvidence-first reportingLanguageENRead in your language
105 reports
- Can AI agents read routine wet-lab evidence? A 100-task benchmark left Astra short of full creditScience & Research
- Can AI formalize mathematics without humans reading every proof? AutoIP shifts the review burden to statementsScience & Research
- Can thermodynamics show what an AI learns about event order? A new measure found a signal—but not causal understandingScience & Research
- Can AI agents conduct publishable physics research? The best FrontierPhysics setup passed 46 of 168 rolloutsScience & Research
- Can AI agents do theoretical computer science research? A 100-task benchmark verified proofs and constructions—but not independent solutions to long-standing conjecturesScience & Research
- Can AI tools review scientific literature reliably? A 102-paper benchmark found access and judgment gapsScience & Research
- Can an AI learn language from one child's everyday speech? Eight home-trained models acquired sounds and thousands of word formsScience & Research
- Can better research handoffs make AI experiments reproducible? An unreviewed benchmark found large gains—but at a substantial construction costScience & Research
- Can hidden AI swarms search science faster? One GPU beat three text-based systems on a 21-problem benchmarkScience & Research
- Did AI scientists beat human research 74% of the time? Only inside ten scaled-down paper environmentsScience & Research
- Does scGPT beat classical models for cell-type annotation? A 965,427-cell benchmark says: it dependsScience & Research
- Can a one-bit change send AI training down another path? A 500,000-network study finds a reproducibility limitScience & Research
- Can AI agents complete end-to-end physics research? The best configuration passed 46 of 168 rolloutsScience & Research
- Can an AI audit clinical networks from weights alone? A follow-up says calculation mattered moreScience & Research
- Can coding agents build trustworthy charts? VisCodeBench finds data and design errors remainScience & Research
- Can optical AI train without recording every state? A fiber-loop experiment cut computationScience & Research
- Can physics-based learning see planets hidden in JWST glare? FLIP improved sensitivity by up to 15-fold—but only on four datasetsScience & Research
- Can plain-language requests design printable metamaterials? One model reached 88.6% end-to-end successScience & Research
- How are scientists using Claude? 207,049 sessions point to analysis and writingScience & Research
- Can a chat interface make drug-repurposing tools easier to use? ChatDRex passed routing tests, not a clinical validationScience & Research
- Can AI agents run a DNA laboratory? AutoDNA completed four workflows—but replication is still narrowScience & Research
- Can AI track changing facts through a long document?Science & Research
- What must Cornell researchers now document when they use AI?Science & Research
- AI agents reproduced 13 of 93 scientific tasks—what failed?Science & Research
- Can AI identify an unknown molecule from 1D NMR alone?Science & Research
- Can sequence-only AI speed drug hit selection without replacing assays?Science & Research
- Do structured textbooks teach AI mathematics better than more examples?Science & Research
- AI agents rejected a result that met accuracy checks. Does that count as doing science?Science & Research
- AI found 12 human proteins that carried mRNA in cells—animals remain untestedScience & Research
- Can AI agents run a radio telescope? A live SETI system still needs an independent testScience & Research
- Do genome AI models predict harmful variants this well? Mixing variant types can inflate the scoreScience & Research
- A local LLM turned chemistry prose into lab steps without retraining—but it did not run the experimentsScience & Research
- Can a scientist describe an image analysis in plain English? CopilotJ completed 84–93% of repeated demo tasksScience & Research
- Did AI discover high-temperature superconductors—or only generate candidates?Science & Research
- One particle model tackled four collider tasks—but its real-data test was narrowScience & Research
- AI agents helped formalise a 15,000-page mathematics proof—but humans still have to inspect the statementScience & Research
- Can a bird-sound model hear threatened fish spawning? It caught only about half the eventsScience & Research
- Can an AI search agent learn to look for evidence, not just get the answer right?Science & Research
- Can hospitals study dementia-drug signals without sharing patient records? A federated method matched pooled estimates—but did not prove treatments workScience & Research
- Do AI research agents turn mixed evidence into a cleaner story than the sources support?Science & Research
- One LLM run can make a prompt look better than it is: repeated tests found randomness dominated some resultsScience & Research
- OpenAI says its models solved three major theory problems. What is actually verified?Science & Research
- Medical reviewers know AI—but 70% in a global survey had not used it for peer reviewScience & Research
- A spinal-cord AI worked across thousands of brain scans—its clinical role remains untestedScience & Research
- Why do diffusion models stop copying and start inventing? A study finds a middle phaseScience & Research
- A protein model reached near-experimental error—then stumbled on engineered mutationsScience & Research
- Can a neural wavefunction obey spin rules by design?Science & Research
- Can AI research agents catch their own scientific shortcuts?Science & Research
- Why did ‘uncertain’ AI choices fail to beat random sampling?Science & Research
- A mammography AI wrote fluent reports—did it get the clinical judgement right?Science & Research
- A biomedical AI cited real papers—but did they support its claims?Science & Research
- AI agents improved 20 scientific benchmarks—but did they do science?Science & Research
- AI papers share far more code and data—but are they really more reproducible?Science & Research
- Giving vision models a pointing tool made their counting visible—not reliably betterScience & Research
- AI reconstructed heart-flow patterns from sparse particles—how close is that to clinical imaging?Science & Research
- An AI model rediscovered a forgotten nuclear shortcut—does it work beyond measured atoms?Science & Research
- Can a written instruction become a working protein?Science & Research
- Do driving agents improve because they remember—or because the prompt changed?Science & Research
- Is AI research narrowing around its biggest successes?Science & Research
- Biology agents could look up protein facts—but geometry exposed a hidden weaknessScience & Research
- Eight vision-language models read charts differently from people—why does that matter?Science & Research
- Can an AI agent audit cancer studies? A 20-paper test found useful triage—and many false alarmsScience & Research
- Keeping patient data local does not make an AI health claim causalScience & Research
- Can a foundation model beat Cox regression on small survival datasets?Science & Research
- LLMs read compact spectra better—but a random forest still wonScience & Research
- OpenAI is offering GPT‑Rosalind to US labs—what is actually free?Science & Research
- Can a saturated AI benchmark still reveal capability?Science & Research
- Did living brain organoids learn to read Braille?Science & Research
- Did training AI for diversity produce better science ideas?Science & Research
- Can AI agents safely assess a historic minaret?Science & Research
- How should dental-imaging AI evidence be judged?Science & Research
- What will CERN's proposed physics foundation model actually do?Science & Research
- Can one AI safely clean up seven kinds of medical images?Science & Research
- Why do medical journals record AI use so differently?Science & Research
- Can AI keep living cells under the nanoscope longer?Science & Research
- Can Korean medical AI be tested on leaked questions?Science & Research
- Can more data fix measurement errors in AI?Science & Research
- Can a general-purpose LLM search crystal compositions?Science & Research
- Can AI choose the next drug-chemistry experiment?Science & Research
- Can AI uncertainty warn of tipping points?Science & Research
- Can METR’s AI time horizons be read literally?Science & Research
- Can researchers rerun clinical AI?Science & Research
- Did the arthritis AI travel between cohorts?Science & Research
- Can a failing AI policy reveal what a tumour model is missing?Science & Research
- Can AI turn knee MRI into a personalised repair scaffold?Science & Research
- Can Tangermeme reveal what genomic AI has learned?Science & Research
- Who is responsible when AI agents do the research?Science & Research
- What did first-day verification find in OpenAI’s mass maths release?Science & Research
- Can AI reveal hidden disease patterns in tissue?Science & Research
- Can an LLM spot what a clinical-trial paper leaves out?Science & Research
- Can AI-controlled droplets make colour tests more reliable?Science & Research
- Can a crash model help dispatchers if it already knows the outcome?Science & Research
- Can a polymer AI predict beyond its simulations?Science & Research
- Why did AI vision miss a human illusion?Science & Research
- Can an automated map keep pace with AI-oncology research?Science & Research
- Can AI design better optimisation rules?Science & Research
- Can AI make a chemical prediction useful in a different laboratory?Science & Research
- Can an AI choose your next experiment before you spend the compute?Science & Research
- People accepted 78% of wrong AI actions when uncertainty stayed hiddenScience & Research
- Can self-supervised AI help neutrino detectors learn from fewer labelled events?Science & Research
- Do AI agents follow their plans? What a benchmark foundScience & Research
- Why do European biotech researchers put AI data mining first for 2030?Science & Research
- Can AI plan under pressure? New Game Arena research tests chess, poker and WerewolfScience & Research
- Microrobot navigation can be trained in minutes in new study; patient use remains untestedScience & Research
- AI-assisted enzyme discovery is promising—but the laboratory still decides what is realScience & Research