Health & Medicine

Anthropic has bought Mithril Medicine, a drug startup with one full-time employee. Founder Cody Tranbarger used Claude to run contracts, logistics and experiments with 11 outside labs, and handed Claude much of the rest of the company's…

Science & Discovery · AI Business & Money · Anthropic

A dermatologist pitted a decade-old, purpose-built image classifier against GPT-6 Astra and Claude Fable 5.1 on 200 confirmed skin biopsies, asking each to pick the correct diagnosis from six options. The old model, fine-tuned for roughly…

OpenAI · Anthropic

Anthropic is opening applications for guest researchers in life sciences, offering academic, nonprofit, and government labs free access to Claude plus 8-12 weeks working alongside Anthropic staff. Target areas include human biology…

Science & Discovery · Anthropic

Quest Diagnostics, the lab company whose patients each year make up about a third of the US adult population, says hackers got into a system doctors use to order tests and look up results, and accessed patient health and identifying…

Cybersecurity · Privacy & Surveillance

Researchers tested four open AI models on Spain's grueling MIR medical residency exam—1,552 questions across eight years—and found the larger models scored well enough to rank for residency placement, with 41 to 43 specialties still open…

Education

Researchers proposed ClinGuard, a governance framework for AI agents coordinating clinical tasks like hospital discharge planning. Rather than letting agents communicate in free text, it routes their proposed actions through an independent…

AI Agents · AI Safety & Risk

OpenAI, PerturbAI, and Scripps Research deployed thousands of AI agents built on a system called GPT-Rosalind to comb through existing data from a whole-brain CRISPR atlas. It covers millions of mouse brain cells tied to specific gene…

Science & Discovery · AI Agents · OpenAI

OpenAI says it wants AI to help scientists through the entire process of discovering and developing new medicines, a journey it says takes 10 to 15 years on average in the US. Now it has bought a company to help get there: Orchestra Bio…

Science & Discovery · AI Business & Money · OpenAI

Scorpions sting more than 1.2 million people a year and kill more than 3,250, mostly children in places with little medical care, yet new treatments draw little investment because they are unlikely to pay back their costs. Anthropic's…

Science & Discovery · Anthropic

Researchers interviewed 21 people living with chronic illness, using prototype chatbots, about how AI could help them explain their illness to others. Researchers describe AI taking on a role once filled mainly by patient communities…

Google showcased its Antigravity AI development platform helping biologists identify a candidate antimicrobial peptide aimed at a bacterium linked to respiratory infections, with the lead now moving to wet-lab testing. Google says the…

Science & Discovery · AI Agents · Google

A review of 45 academic papers on AI mood-tracking and emotion-sensing tools—apps that use chatbots to help people log and reflect on their feelings—found most systems share the same blind spots. They lean on clunky, multi-step text…

Kids, Teens & Mental Health

Researchers built a prototype AI tool to help hospitals schedule nurses, co-designing it with healthcare professionals, then tested it with eye tracking and questionnaires. The twist: telling users how confident the AI was in its forecasts…

AI Adoption

A new academic study argues that AI disease-surveillance tools fail in poorer countries not because the technology is inaccurate, but because of deeper structural problems: whether the data reflects local realities, who benefits from the…

A study interviewed 13 researchers, regulators, and other stakeholders involved in deciding whether animal drug trial results justify moving to human trials. It found they want AI to help find, sort, and pull relevant evidence from…

Science & Discovery

Researchers deploying diagnostic imaging AI at six hospitals report the real bottleneck isn't model accuracy—it's plumbing: routing scans to the right AI tool, displaying results usefully, and auditing what ran. Their open-source platform…

Researchers built EMooly, a tablet game combining augmented reality and generative AI to help autistic children practice recognizing emotions, with caregivers participating alongside the child. The study was published in an ACM journal…

Kids, Teens & Mental Health

A study of 26 exercise physiologists testing an AI tool that drafts physical-activity plans for cardiovascular patients found no overall boost in speed, confidence, or plan quality. But the results split by user: clinicians who struggled…

AI tools that listen in on doctor visits and auto-generate clinical notes—already spreading in U.S. hospitals—hit a data problem in India, a new study finds. Researchers surveyed available training datasets and interviewed five…

A new study finds that medical AI models can develop a hidden "memorisation bias": if a patient's past health records were part of a model's training data, that history quietly skews the model's later predictions for the same patient. When…

Privacy & Surveillance

Bioengineer César de la Fuente's lab is using AI—custom deep-learning models alongside ChatGPT and OpenAI's coding tool Codex—to scan genomes of living and extinct organisms for molecules that could become new antibiotics. The team says…

Science & Discovery · OpenAI

A new study tested whether GPT-4 could help hospitals flag concerning Emergency Department revisits—cases where a patient returns soon after discharge, potentially signaling a missed diagnosis. Asked directly, GPT-4 flagged 94% of 99…

A new study pitted a clinical AI system called Doctorina against eight physicians and four leading language models on 150 simulated Polish primary-care visits. Doctorina correctly identified the top diagnosis 82% of the time versus 57% for…

A research team built VERDICT, an AI agent for matching patients to clinical trials that translates eligibility rules and constraints into formal logic, then hands the decision off to a mathematical solver rather than letting the language…

AI Agents

OpenAI added two integrations to ChatGPT for Healthcare: a connection to Epic's electronic health record system and a plugin pulling from nine official public health databases, including PubMed, ClinicalTrials.gov and CMS coverage data.…

OpenAI

A clinician-authored study details TrialGPT 2.0, a system that matches cancer patients to clinical trials, tested across government, academic, and NIH referral settings—not just in a lab. Across 288 retrospective cases, it surfaced at…

A study of 107 psychiatric interviews found that clinicians' judgments of how a patient experienced a session—rated afterward by the interviewer—only loosely matched what patients themselves reported (a correlation of 0.365 out of a…

Kids, Teens & Mental Health

An audit of three commercial AI medical scribes—tools that listen to doctor-patient visits and draft clinical notes—found verified errors in 31.3% of 565 notes, concentrated in allergy and medication details, invented patient identities…

A study tested 11 leading AI models on 208 rare-disease ethics scenarios requiring tradeoffs between competing medical principles—fairness, patient benefit, avoiding harm, and autonomy. Every model defaulted to strict equal resource…

AI Safety & Risk

Researchers built HealthBench-Psych, a mental-health-specific slice of OpenAI's HealthBench benchmark, by filtering its 5,000 physician-graded conversations down to 610 focused on psychological support—then validated the selection through…

Kids, Teens & Mental Health · OpenAI

A new study finds people trust health information more when it comes from ChatGPT than from Google, even when the underlying content is similar. A second experiment found trust also shifts depending on how the same AI-generated answer is…

OpenAI · Google

Researchers built a new AI training method and benchmark aimed at a specific problem: translating dense medical reports—radiology notes, lab results—into language patients can actually understand. The framework, called G-CARL, checks…

Researchers built an AI model that predicts how heart patients will fare after atrial fibrillation ablation—a common procedure to correct irregular heartbeats—by tracking medication changes, repeat procedures, and vital signs over time as…

A medical AI system built specifically for clinical knowledge retrieval in India and other lower-income health settings held its own against, and often beat, general-purpose frontier models on HealthBench, a benchmark scored using…

Google researchers tested AMIE (Video), a Gemini-based system that conducts clinical consultations over live video, against real primary care physicians in a simulated exam with 30 doctors, 15 trained patient actors, and 100 case…

Google

OpenAI rolled out updates to GPT-5.6 in ChatGPT: Plus and Pro subscribers get Sol with a new slider to dial reasoning effort up or down, while Free users now default to Luna, which includes unlimited text chats and a "Think" button for…

AI Business & Money · OpenAI

For the first time, scientists have used AI to design entirely new viruses that actually function. Researchers at the Arc Institute, a nonprofit lab in Palo Alto, used two "genome language models"—Evo 1 and Evo 2, trained on millions of…

Science & Discovery · AI Safety & Risk

Researchers unveiled CoPlan, a prototype AI system for building elder-care plans that keeps human caregivers in charge. Rather than issuing recommendations outright, it uses multiple AI agents to generate candidate interventions along with…

AI Agents

A study tested three AI-assisted writing tools designed to help ordinary social media users craft rebuttals to vaccine-skeptical posts. Users who got AI help—whether drafting from scratch or revising their own text—produced responses rated…

Deepfakes & Misinformation

Researchers unveiled RemiAssist, an AI tool built to help therapists—not patients—run photo-based reminiscence therapy for people with dementia. The system organizes life-event photos into a structured map for session planning and offers…

Digital phenotyping tools infer mental health conditions from behavioral data like spending patterns and phone use, raising obvious privacy and consent concerns. Researchers built a formal system that encodes ethical rules—like when data…

Privacy & Surveillance

Researchers built SIT-CARE, an AI tool that helps clinicians diagnose autism spectrum condition by flagging clinically relevant nonverbal behaviors and offering recommendations during assessment. The design grew out of interviews…

OpenAI is rolling out a Health feature in ChatGPT for logged-in U.S. users 18 and up, letting people connect Apple Health data and medical records so the chatbot can reference that information across any conversation, not just a dedicated…

Privacy & Surveillance · OpenAI

Researchers built MedGame, a system that converts static medical case studies into interactive branching-story games, along with a benchmark of 5,000 clinical cases to test how well AI models generate these narratives. The team reports…

Education

A study based on interviews with 14 researchers and practitioners across five countries examined passive sensing systems—apps and wearables that infer depression, stress, or cognitive load from phone and sensor data. The researchers found…

Privacy & Surveillance

Kaiser Permanente nurses who handle patient advice calls say AI-powered surveillance is undermining their care. Seven current and former nurses told CalMatters that systems rating their empathy, tone, and predicted productivity pressure…

AI & Jobs · Privacy & Surveillance

Researchers have proposed a framework for annotating depression symptoms in clinical data that pairs AI labeling with psychiatrist oversight. The system uses a three-stage process aligned with DSM-5-TR diagnostic criteria: selecting…

Kids, Teens & Mental Health

Researchers developed a framework for detecting health misinformation in low-resource languages—those with limited AI training data—using Bangla as a test case. The approach combines smaller, efficient language models with cultural…

Deepfakes & Misinformation · Microsoft

A large-scale field experiment at a Chinese hospital tested what happens when patients consult an AI chatbot before seeing their doctor. The preregistered study found that AI advice nudging patients away from certain…

AI models can match physicians' scoring accuracy on medical questions but lack a crucial clinical instinct: knowing when to say "I'm not sure." Researchers created MedQADE, the first open-response clinical benchmark in German, with 3,800…

Google

Researchers developed a reinforcement learning method called MRPO that fixes a specific problem in medical AI: when diagnostic reasoning goes wrong early, every subsequent step compounds the error. Their approach penalizes early mistakes…

A patient with a diagnosed shoulder injury decided to get a second opinion—from an AI. He fed his raw MRI files to Claude Code running Opus 4.8 and asked it to assess his rotator cuff. The result: the AI contradicted his doctor's diagnosis…

Anthropic · OpenAI

Surgeons want AI as a copilot, not an autopilot. A study of 17 surgeons designing an AI interface for gallbladder surgery found near-unanimous agreement (16/17) that AI should support decisions, not make them. Experienced surgeons…

A study of over 15,000 user reviews across 59 AI healthcare chatbot apps found three recurring breakdown categories: access barriers and service unreliability, poor user experience and interaction quality, and billing and customer support…

Privacy & Surveillance

Researchers released MedSWFlow, an open-source framework that uses large language models to draft medical social work case plans. The system walks through six stages—assessment, problem analysis, goal setting, intervention planning, risk…

A six-week study testing WhatsApp fitness chatbots found a counterintuitive split: users rated the least empathetic bot as more engaging and useful, yet the high-empathy versions actually drove more behavior change—larger increases in…

OpenAI says GPT-5.5 Instant now matches its most advanced reasoning models on health questions, a capability it's rolling out free to all ChatGPT users. The company reports a 71% drop in flagged factual errors over the past two months…

OpenAI

Medical imaging is one of the biggest bottlenecks in stretched health systems: MRI machines cost millions, scans run hundreds to thousands of dollars, and wait lists stretch for months. Midjourney — the AI image generator — says it has a…

A consortium of 173 scientists has released LifeSciBench, a benchmark designed to test whether AI systems can handle actual life science research work—not just answer textbook biology questions. The 750 tasks span drug discovery, genomics…

Science & Discovery

OpenAI connected GPT-5.4 to Molecule.one's robotic chemistry lab and let it autonomously research ways to improve Chan-Lam coupling—a reaction used in drug development that's notoriously finicky. The AI independently identified a difficult…

Science & Discovery · OpenAI

Researchers have developed a taxonomy mapping Alzheimer's and dementia caregivers' mental health needs to technology-based interventions. The framework, built from literature review and caregiver interviews, identifies gaps in current tech…

Researchers studying AI-generated support for Alzheimer's caregivers found a troubling pattern: when prompted to act as peer supporters, models like GPT-4o-mini and LLaMA produce warm, relatable responses that mimic lived experience—but…

Kids, Teens & Mental Health · OpenAI · Meta

A qualitative study interviewed 25 U.S. emergency medical technicians and paramedics about AI in their workflows—and found deep skepticism. Clinicians worried that AI tools could disrupt the rapid, distributed teamwork that emergency…

AI Adoption

A working paper published on NBER examines how medical liability law should work when AI systems act as doctors rather than decision-support tools. The paper argues liability design depends heavily on how well medical records document what…

Courts & Lawsuits

Who actually benefits when AI assists an expert? A new replication suggests the gains are far from uniform. Building on a 2025 framework from Andrew Caplin, David Deming and colleagues, Daniel Martin tested whether its predictions hold for…

AI Adoption

Medical AI systems that ace licensing exams may still be dangerously unreliable. A new benchmark called MedMisBench tested whether LLMs can maintain correct medical judgment when fed misleading information—and found they frequently can't.…

AI Safety & Risk

Researchers created a framework for clinical AI that separates two types of uncertainty—randomness in the data itself versus gaps in the model's knowledge—and used it to audit for algorithmic bias across patient groups. Testing on 1,000…

A new research framework called Ouvia evaluated how well speech translation actually works for real conversations—and the findings are sobering. Testing four translation systems across 1,750+ healthcare and everyday interactions between…

Researchers created PhysDox, a benchmark testing whether LLMs can spot physically impossible steps in biomedical sensing protocols—like detecting when a heart monitor procedure violates basic physics. Even top models peaked at just 53%…

OpenAI upgraded GPT-Rosalind, its enterprise AI for life sciences, integrating GPT-5.5's agentic coding and tool-use abilities with domain expertise in drug discovery, medicinal chemistry, and genomics. The company claims broad performance…

Science & Discovery · OpenAI

Researchers have published a paper describing CARE-link, an open-source platform that uses LLMs to help clinicians manage gestational diabetes patients remotely. The system aggregates patient data, provides clinical decision support, and…

Open-Weight Models

Researchers have proposed using generative AI to create abstract "visual summaries" of daily activities for elderly care—offering adult children awareness of aging parents without invasive camera monitoring. The concept would translate raw…

Privacy & Surveillance

A new medical benchmark using 1,638 standardized patient cases finds current AI models fall well short of clinical reliability. GPT-4.5, the top performer, completed just 60.4% of expert-defined diagnostic steps; medically specialized…

Researchers tested whether AI-generated summaries of health tracking data could help families remotely monitor older relatives. The key finding: summaries that explain 'how' and 'why'—not just 'what' happened—performed significantly…

A benchmark called ClinEnv tests AI models as attending physicians, requiring them to gather information incrementally and make irreversible decisions—mimicking real hospital admissions rather than answering static medical questions. The…

Researchers tested how assigning different "support roles" to AI chatbots—informer, coach, relator, listener—affects safety when answering real caregiver questions about Alzheimer's and dementia. The study used 5,000 actual queries from…

AI Safety & Risk

Boston Children's Hospital reports that AI has become core infrastructure across its operations, with more than a third of employees now using it daily. The hospital says it has diagnosed over 40 rare conditions that previously went…

AI Adoption · OpenAI

OpenAI launched Rosalind Biodefense, an initiative giving select U.S. government agencies and allied partners access to GPT-Rosalind, the company's specialized reasoning model for life sciences. The program aims to help trusted developers…

Military & National Security · Science & Discovery · OpenAI

New research reveals a significant gap in how AI performs medical diagnosis: LLMs consistently score lower on diagnostic accuracy when given structured electronic health record data (the FHIR format hospitals actually use) compared to…

A study of 192 participants found that how AI systems display uncertainty affects whether users verify the information. When uncertainty was shown at the word level (highlighting specific uncertain terms), users agreed more with the AI.…

AI Adoption

A position paper argues that healthcare AI benchmarks fail to predict real-world performance not because they're poorly designed, but because they can't account for how humans actually interact with these systems. The researchers analyzed…

A large-scale audit of over 6,200 medical chatbots built on GPT found alarming safety gaps: 25-30% showed low factual accuracy, up to 54% violated operational safety thresholds, and 57% of those with action capabilities (like booking or…

AI Safety & Risk · Privacy & Surveillance · OpenAI

AdventHealth, a nine-state hospital system, is rolling out ChatGPT for Healthcare organization-wide after piloting ChatGPT Enterprise. The system claims an 80% reduction in time spent on administrative tasks, with physician advisors who…

AI Adoption · OpenAI

MIT biologists are using Google's Co-Scientist AI tool to hunt for genes that could reverse cellular aging. After analyzing tens of thousands of papers, the system proposed over 20 novel genetic factors worth investigating—and lab tests…

Science & Discovery · Google

Researchers developed TimeSRL, a framework that converts raw behavioral data (like sleep patterns or phone usage) into plain-language descriptions before making predictions—a "semantic bottleneck" approach. In mental health prediction…

Kids, Teens & Mental Health

Researchers developed CandorMD, an AI-powered audio simulation system designed to train doctors in one of medicine's most difficult conversations: telling patients about medical errors. The system provides real-time practice scenarios and…

Researchers developed PromptRad, a method for automatically labeling radiology reports that matches GPT-4's performance using a far smaller model—and needs only 32 labeled examples to train. The approach works by reformulating the labeling…

A study examining AlphaFold2's impact since its 2021 release found a surprising split: basic research on previously unstudied proteins jumped 15-40%, but experimental lab work determining protein structures continued at nearly the same…

Science & Discovery · Google

A fairness audit of five major LLMs used for emergency department triage found all failed a pre-set bias threshold, with gender-related decision flip rates ranging from 9.9% to 43.8% when patient gender was swapped in otherwise identical…

New research challenges a core assumption in healthcare AI: bigger models aren't automatically safer. Researchers tested 34 clinical LLMs and found that accuracy and safety follow different scaling laws. The key factor isn't model…

AI Safety & Risk

A Google-backed study of nearly 14,000 Fitbit app users found that AI agents conducting symptom interviews produced more accurate differential diagnoses than independent clinicians working from the same conversation transcripts. The AI was…

Google

Google DeepMind unveiled an AI co-clinician research initiative exploring how AI could work alongside physicians—interacting directly with patients while doctors retain clinical authority. In head-to-head evaluations, physicians preferred…

Google

A study testing 15 major AI chatbots on psychiatric triage found they rarely miss genuine emergencies—correctly identifying cases requiring urgent medical attention 94% of the time—but consistently err toward caution on less severe cases.…

Kids, Teens & Mental Health

Researchers proposed Bezier Trajectory Matching (BTM) for dataset condensation—the process of creating small synthetic datasets that can train models nearly as well as full-sized ones. Instead of using standard training trajectories to…

Researchers have developed DiCE, an AI framework for analyzing capsule endoscopy videos—the pill-sized cameras patients swallow for GI tract imaging. The challenge: these videos run 8+ hours, and physicians must manually review thousands…

OpenAI has launched ChatGPT for Clinicians, a free version of ChatGPT available to verified U.S. physicians, nurse practitioners, and pharmacists. The company says the specialized tool supports clinical care, documentation, and research…

OpenAI

Researchers built Apollo, a medical AI trained on 25 billion records from 7.2 million patients at a major US hospital system, spanning three decades and 28 types of medical data—from lab results to imaging to clinical notes. The model…

Researchers released IDOBE, a benchmark dataset for epidemic forecasting that compiles over a century of disease surveillance data across U.S. states and global locations, covering 13 diseases with more than 10,000 outbreak segments. The…

Researchers introduced DLink, a framework for compressing large AI models that interpret brain signals (EEG) so they can run on small, wearable devices. The approach uses knowledge distillation—training a compact 'student' model to mimic a…

OpenAI announced GPT-Rosalind, a reasoning model built specifically for life sciences work—drug discovery, genomics, protein analysis, and scientific research workflows. The company claims the model is designed to accelerate research in…

Science & Discovery · OpenAI

Researchers developed RadAgent, an AI system that generates chest CT reports through a step-by-step process with a fully transparent reasoning trail. Unlike black-box AI models, RadAgent shows clinicians exactly how it reached each…

Researchers developed LAMAE, a foundation model designed specifically for echocardiography—the ultrasound imaging used in cardiac diagnostics. The key finding: AI models trained on adult heart scans can transfer effectively to pediatric…

Researchers tested whether AI can detect ambivalence and hesitancy in video—emotional signals that could help digital health apps respond more appropriately to uncertain users. They compared supervised learning, personalization techniques…

A new benchmark reveals that AI vision models perform poorly at judging whether medical procedures are being done correctly—even when their overall scores suggest otherwise. SiMing-Bench tested leading multimodal AI systems on real…

Researchers have released open-source code and data for classifying patient activity levels—lying, sitting, standing, walking, jogging—using accelerometer data from wearable devices. The approach, tested on 23 healthy subjects, achieved an…

Researchers have developed ECHO, an AI system for generating chest X-ray reports that takes a fundamentally different approach from current methods. Instead of producing text word-by-word like ChatGPT-style models, ECHO uses a…

Researchers developed a brain-decoding AI that can interpret what someone is seeing from fMRI scans—without needing to be trained on that specific person's brain data. The system uses in-context learning (similar to how ChatGPT adapts to…

Privacy & Surveillance

Medical imaging researchers developed a framework that mathematically separates what MRI and PET scans can each reveal about prostate cancer. By decomposing the imaging data into orthogonal components, the team found that PET scans capture…

Researchers released a dataset of 8,800 synthetic doctor-patient conversations—1,300 hours of audio with matching clinical notes—designed to train AI systems that summarize long medical appointments. The conversations were generated…

Researchers built a speech recognition system specifically for gastrointestinal endoscopy procedures, allowing doctors to dictate findings hands-free during scopes. The system, called EndoASR, dramatically improved medical terminology…

Researchers proposed a framework classifying quantum biosensors into four generations based on how they harness quantum physics. The first two generations use basic quantum properties for measurement; the third achieves far greater…

Researchers found that deep learning models trained entirely on synthetic fractal images—mathematical patterns with no medical data—can reconstruct real-time cardiac MRI scans as well as models trained on actual heart imaging. In tests on…

Sid Sijbrandij, GitLab's founder, is fighting osteosarcoma in his spine by treating his cancer like a startup problem—running treatments in parallel, pursuing maximum diagnostics, and launching companies through evenone.ventures to scale…

Researchers have developed a reinforcement learning technique that trains language models to generate multiple candidate answers with confidence estimates in a single pass, rather than fixating on one "best" answer. The approach addresses…

Researchers tested DeepFAN, a transformer-based AI model trained on over 10,000 pathology-confirmed lung nodules, in a clinical trial across three Chinese medical institutions. When 12 junior radiologists used the system to evaluate 400 CT…

Researchers developed RAVEN, a foundation model trained on electronic health records from over one million patients that can predict disease incidence without task-specific fine-tuning. The model, which learns to forecast clinical events…

OpenAI announced its foundation arm will invest at least $1 billion in philanthropic initiatives across four areas: curing diseases, expanding economic opportunity, building AI resilience, and community programs. The commitment comes as…

AI Business & Money · OpenAI

Researchers developed a technique to detect dysarthria—a speech impairment common in Parkinson's disease—across multiple languages using a single AI model. The method aligns speech representations from different languages into a common…

Stanford researchers developed MARCUS, a specialized AI system for interpreting cardiac imaging—ECGs, echocardiograms, and cardiac MRIs. The system uses coordinated expert models trained on 13.5 million medical images. In testing, MARCUS…

OpenAI · Google

Researchers released ODySSeI, an open-source framework that automatically detects, outlines, and assesses severity of blockages in coronary angiography images—the X-ray videos cardiologists use to spot heart disease. Trained on data from…

Open-Weight Models

Researchers have released HEALIX, the first publicly available dataset for detecting patient health literacy levels from clinical notes. The dataset contains 589 annotated clinical notes labeled as low, normal, or high health literacy…

Google is partnering with Australian healthcare organizations to deploy its Population Health AI in rural communities where residents are 60% more likely to die from heart disease than those in cities. The $1 million AUD initiative pairs…

Google

Researchers from academic institutions, a health tech company, and a hospital built and tested a phone-based maternal health chatbot for India that handles multilingual queries—including the code-mixed speech common in real conversations.…

AI Safety & Risk

Researchers developed SAW (Surgical Action World), a video generation model that creates realistic laparoscopic surgery footage from simple inputs: text prompts, reference images, tissue maps, and tool movement paths. Trained on 12,044…

A study tested seven open-source large language models on Japanese pathology report tasks—structured diagnosis generation, information extraction, typo correction, and explanatory text. Researchers found these models can help in limited…

Open-Weight Models

Google is partnering with Australian health organizations to deploy AI-powered screening in rural communities, where residents are 60% more likely to die from heart disease than metropolitan populations. The $1 million AUD initiative uses…

Google

Researchers developed HeartAgent, an AI system designed to help cardiologists with differential diagnosis. The system combines specialized sub-agents and curated medical data to generate diagnoses with transparent reasoning and verifiable…

Researchers introduced PathMem, an AI framework designed to help pathology models reason more like human diagnosticians. The system organizes medical knowledge into structured long-term memory and dynamically retrieves relevant information…

Researchers trained an AI system on over 45,000 fetal ultrasound images from 22 hospitals to detect orofacial clefts—birth defects notoriously difficult to spot on prenatal scans. The system reportedly matches senior radiologist…

Researchers have developed AIRT, a system that generates radiation treatment plans for prostate cancer directly from CT scans in under one second—compared to several minutes for current automated tools. Trained on more than 10,000 cases…

A study testing 34 large language models on 169 expert radiology questions found that giving all models the same structured evidence—via a retrieval-augmented reasoning pipeline—dramatically reduced how much their answers varied. Median…

Researchers released Med-V1, a family of 3-billion-parameter language models designed specifically for biomedical fact-checking—detecting hallucinations in AI-generated medical content and verifying clinical claims against source…

Open-Weight Models

Researchers developed a new AI classification method to assess stroke risk from carotid artery ultrasound images, analyzing 500 plaques from the CREST-2 multi-center clinical trial. The approach combines machine learning with…

Researchers released ClinConsensus, a benchmark of 2,500 open-ended clinical cases in Chinese designed to evaluate medical LLMs more rigorously than multiple-choice tests allow. The cases span 36 medical specialties and 12 clinical task…

A team of researchers has built OpenRad, a curated repository of roughly 1,700 open-access radiology AI models drawn from peer-reviewed papers and preprints through December 2025. The database spans all major imaging types—CT, MRI, X-ray…

Researchers developed MedCLIPSeg, a framework that adapts OpenAI's CLIP model for medical image segmentation—the task of identifying and outlining specific structures in medical scans. The approach claims to outperform prior methods across…

OpenAI

Researchers have proposed ODEBRAIN, a framework for analyzing brain activity from EEG data that models neural dynamics as continuous rather than discrete time steps. The approach combines spatial, temporal, and frequency information from…

Science & Discovery

Researchers have developed ColoDiff, a framework that generates synthetic colonoscopy videos for training medical AI systems. The approach addresses a persistent problem in healthcare AI: getting enough training data when real medical…

Researchers developed an AI framework that can detect coronary artery calcium—a key cardiovascular risk marker—from routine chest CT scans, not just the specialized cardiac scans typically required. The system, called CARD-ViT, trained…

Researchers have developed LUMEN, a training framework designed to help AI models interpret chest X-rays taken over time rather than as isolated snapshots. The system uses a question-and-answer interface to handle both diagnosis (what's…

Researchers developed PVminer, an NLP framework designed to extract structured insights from patient-generated text—secure messages, surveys, and interviews. The tool combines specialized language models with topic analysis to identify…

UC San Diego researchers unveiled HiAER-Spike, a neuromorphic computing platform that processes information using "spiking" neural networks—systems that mimic how biological neurons fire in bursts rather than continuous signals. The…

Researchers report a hybrid quantum-classical approach to satellite image classification that achieved 87% accuracy on IBM quantum processors—a 2-3 percentage point improvement over classical methods including ResNet50 with transfer…

Researchers have developed MolHIT, a new AI framework for generating molecular structures that claims state-of-the-art performance on the MOSES benchmark—a standard test suite for molecule generation. The system uses a hierarchical…

Science & Discovery

Researchers developed a new AI framework for diagnosing Alzheimer's disease and Lewy body dementia by analyzing brain cortical folding patterns. The approach uses a probability-based method to classify brain networks without requiring the…

Researchers have developed CAMEL, which they describe as the first AI model combining ECG signal analysis with language understanding to predict future cardiac events—not just diagnose current conditions. The model analyzes longer…

OpenAI published a case study of a family that used ChatGPT to prepare for cancer treatment decisions for their son, framing the chatbot as a complement to—not replacement for—his medical team. The piece describes using it to understand…

OpenAI

A study testing 11 large language models on PTSD severity assessment found that context matters enormously: LLMs performed best when given detailed clinical definitions and background information, not just raw patient narratives. More…

Kids, Teens & Mental Health

OpenAI published a case study of a family who used ChatGPT to help process information and prepare questions during their son's cancer treatment. The family used the tool alongside their medical team—not as a replacement—to understand…

OpenAI