AI* and healthcare/medicine

Doctor forced to apologise after AI makes 'scary' error about illegal drugs

Artificial Intelligence is listening to many people's medical appointments, but it is not always hearing things correctly.

AI creates first synthetic viruses

Genomic language model has huge potential to redesign organisms such as bacteria

Generative AI comes to gene editing

Profluent releases AI-designed gene editor, OpenCRISPR-1

Structure and evolution-guided design of minimal RNA-guided nucleases

The design of RNA-guided nucleases with properties not limited by evolution can expand programmable genome-editing capabilities. However, generating diverse multidomain proteins with robust enzymatic properties remains challenging. Here, we use a protein design strategy that couples a structure-guided inverse-folding model with evolution-informed residue constraints to generate active, divergent variants of TnpB, a minimal CRISPR-Cas12–like nuclease, termed SynTnpBs. High-throughput screening of artificial intelligence–generated variants yielded editors that retained or exceeded wild-type activity in bacterial, plant, and human cells. Cryo–electron microscopy–based structure determination of the most divergent variant revealed stabilizing contacts in the RNA–DNA interfaces across conformations, demonstrating the design potential of this approach. Together, these results establish a strategy for creating non-natural RNA-guided nucleases and conformationally active nucleic acid binders, enlarging the designable protein space. , Editor’s summary Enzyme design and engineering are challenging in part because there are few ways to improve catalytic activity but many ways to impair it. Evolution-aware approaches can leverage information from our collection of known natural sequences to guide the generation of diverse engineered enzymes that are more likely to retain function. Skopintsev et al . demonstrated engineering of an RNA-guided nuclease using inverse protein-folding models. Screening of candidates nominated by their approach revealed improvement of genome-editing activity. An experimental structure revealed how conformational dynamics are changed in the engineered enzymes, stabilizing the key RNA-DNA interface. —Michael A. Funk

DeepMind releases structure predictions for nearly every known protein

Database powered by AlphaFold algorithm now boasts predicted structures for over 200 million proteins

AI-designed nucleases build on nature’s design

Researchers used AI to create variants of a CRISPR-Cas12-like nuclease, some of which show increased editing activity

Public use of a generalist LLM chatbot for health queries

Here we analyse over 500,000 de-identified health-related conversations with Microsoft Copilot from January 2026 to characterize what people ask conversational artificial intelligence (AI) about health. We apply a hierarchical intent taxonomy of 12 primary categories using privacy-preserving large language model-based classification validated against expert human annotation and use topic clustering for prevalent themes within each intent. We then characterize the intents and topics behind health queries, identify who they are about, and analyse how usage varies by device and time of day. Nearly one in five conversations involves personal symptom assessment or condition discussion, and the dominant general information category is also concentrated on specific treatments and conditions, suggesting that this is a lower bound on personal health intent. One in seven of these personal health queries concerns someone other than the user, suggesting that conversational AI can also be a caregiving tool. Personal queries increase markedly in the evening and nighttime hours, when traditional healthcare is most limited. Usage diverges sharply by device: mobile concentrates on personal health concerns, while desktop is dominated by professional and academic work. A substantial share of queries focuses on navigating healthcare systems. These patterns have direct implications for platform-specific design, safety considerations and the responsible development of health AI.

Kaiser nurses say surveillance of them is undermining healthcare – The Markup

Call center nurses at Kaiser Permanente said workplace surveillance tools and AI prioritize speed and cost savings over quality and safety.

Your medical provider might be recording your mental health care visits – The Markup

Mental health providers are increasingly using AI technology to record conversations, raising privacy concerns among patients and practitioners.

KFF Tracking Poll on Health Information and Trust: Use of AI For Health Information and Advice | KFF

This poll finds that about as many adults are turning to AI for health information as social media, with health care costs and access driving many users, particularly younger users.

KFF Tracking Poll on Health Information and Trust: Use of Social Media and AI For Health Information and Advice | KFF

This poll finds that about 3 in 10 adults turn to social media for health information and advice at least monthly. Community connection and the need for immediate answers are the top reasons why people are turning to these tools. Slim majorities of those who use social media for health are confident they can tell what is true, and relatively few take steps to check the information they receive.

Frequent AI chatbot users more likely to believe anti-vaccine myths, poll finds

Poll finds use of AI tools for health advice is correlated with belief in vaccine falsehoods, such as shots causing autism

The AI Chemist: To be trustworthy, LLMs need to show their work

Good scientists reveal how they do their experiments and report their results; so should any machine-driven research

Office of the Auditor General of Ontario

Home / Special Reports / Use of Artificial Intelligence in the Ontario Government

Your doctor’s AI notetaker may be making things up, Ontario audit finds

Made-up therapy referrals, incorrect prescriptions among the common mistakes.

Preparing Physicians for the Clinical Algorithm Era | NEJM

The U.S. government recently took steps to ensure that clinical decision support algorithms are safe for clinical use. The next and larger step will be teaching physicians how to use the algorithms ...

AI model detects very early normally ‘invisible’ tissue changes of pancreatic cancer

An AI model (REDMOD) can pick up the very early subtle tissue changes of pancreatic ductal adenocarcinoma, the most common form of pancreatic cancer, which conventional imaging and the human eye find difficult to detect, finds research published online in the journal Gut.  As such, it offers the potential to shift an all too common late stage, terminal disease diagnosis to one that is at an early stage (stage 0) and treatable, say the researchers.  While REDMOD was more accurate than experienced radiologists, it requires testing in high risk patients, defined as those with unexpected weight loss and newly diagnosed diabetes, before it can be widely used in clinical practice, they add.