JMIR AI

Large language model–based chatbots (LLM-CBs) are increasingly used as mental health support tools. Risks and harms are discussed especially for unsupervised use and for vulnerable groups. A naturalistic characterization of the use of LLM-CBs by patients with mental health disorders is currently lacking.

Adverse event reports referring to the same case but treated as independent can negatively impact statistical analysis and may mislead clinical assessment. Pharmacovigilance relies on large databases of adverse event reports to discover potential new causal associations. Their size necessitates computational methods to identify duplicates at scale. The current state-of-the-art is statistical record linkage which outperforms rule-based approaches. In particular, vigiMatch is in routine use for VigiBase, the World Health Organization global database of adverse event reports, and represents the first statistical duplicate detection approach in pharmacovigilance deployed at scale. Originally developed for both drugs and vaccines, its application to vaccines has been limited due to inconsistent performance across countries.

AI technologies are currently experiencing significant growth in development and use in health care. The AI model lifecycle includes several stages, all of which should be adequately described. Reporting standards are commonly used for this purpose. However, the plain-text structure of these standards prevents transparent machine-readability of this information.

Symptom detection is essential in global disease surveillance to detect potential outbreaks, as symptoms are the first observable signs of infection. To reflect real-time ground truth conditions during pandemics, social media has emerged as a valuable data source. Moreover, effective digital disease surveillance systems must operate across diverse linguistic settings, and large language models (LLMs) have been shown to perform inconsistently across languages, tending to have lower performance in low-resource languages. While multilingual approaches have been explored in various health-related natural language processing tasks, a critical gap remains in understanding whether LLM-based symptom detection can perform consistently across languages for global disease surveillance. Southeast Asia demonstrates this challenge, combining diverse languages and the potential for emerging infectious disease outbreaks, making it a case for evaluating how multilingual performance disparities manifest in symptom detection.

Independent audits of medical large language models have concentrated on models that answer unsafely rather than those that decline safe questions. In June 2026, Anthropic released Claude Fable 5 with a safeguard that detects requests related to cybersecurity, biology, or chemistry and reroutes them to a fallback model, Claude Opus 4.8. The developer states that this safeguard is deliberately conservative and affects fewer than 5% of sessions. Whether that figure holds for consumer health questions, many containing dense biomedical vocabulary, was unknown.

Personalizing physical activity recommendations for older adults requires understanding not only which dimensions of physical activity and sedentary behaviors (24-h movement behaviors) influence health outcomes but also when, within an individual’s everyday life, these dimensions are most relevant. Current observational and interventional approaches rarely capture the temporal dynamics linking everyday patterns of 24-hour movement behaviors to cognitive and mental health trajectories, 2 key determinants of healthy aging.

Conversational AI systems are increasingly being deployed in health care for clinical decision support, but their performance varies substantially across patient communication styles, health literacy levels, and behavioral patterns. Static benchmarks cannot capture multiturn dynamics through which this variation compounds, and no current evaluation framework implements structured AI risk management guidance for conversational health care AI. The result is a structural risk: AI systems may perform well in aggregate while failing disproportionately for the populations they are intended to help.

Quality-of-life (QoL) questionnaires are an established instrument designed to assess overall well-being and QoL of patients. They are important in predicting the outcome of the disease and understanding the needs of individual patients. However, their repeated collection imposes a substantial burden on both patients and clinical professionals. Many patients seek emotional support and mutual exchange in online communities for peer support, where they frequently share detailed descriptions of symptoms and treatment experiences, addressing topics covered in QoL questionnaires. The emergence of large language models (LLMs) uncovers potential for automatic extraction of relevant QoL information from patient-generated text.

Oral disease and malnutrition are common and closely linked problems in long-term care (LTC). Monthly weights, occasional diet reviews, and infrequent dental assessments can miss gradual decline. Recent tools, including computer-vision meal-intake estimation and smartphone-based gingival screening, create an opportunity for more timely clinical monitoring when limited data capture is embedded into routine care. This viewpoint proposes a nursing-led policy framework for AI-enabled oral and nutrition risk detection in nursing homes. The framework emphasizes protected oral health and nutrition champion roles, a practical 48-hour bedside assessment standard for high-priority operational alerts, standards-based electronic health record (EHR) integration, and prevention-oriented escalation pathways. We clarify that the proposed approach does not require a single black-box AI risk score. Instead, AI-derived measurements, such as estimated intake, plate-waste ratio, deviation from baseline, and image-based oral findings, can be combined with weight trends, EHR data, operational thresholds, and nurse review. The playbook specifies staged rollout, staff-facing alert outputs, fidelity checks for data capture, fallback documentation options for facilities with lower digital maturity, and key performance indicators for clinical outcomes, workflow burden, equity, and cost. Ethical safeguards include layered consent, minimum-necessary capture, opt-out recording, explainability for residents and proxies, and subgroup monitoring. AI-enabled clinical monitoring can support earlier action in LTC only if it is embedded in nursing workflows, auditable documentation, and accountable governance. Prospective, co-designed implementation studies are needed to test feasibility, workload, effectiveness, and equity across diverse LTC settings.

Human-autonomy teaming (HAT) has the potential to reshape surgical practice by fostering true partnership between surgeons and intelligent systems. To achieve this, AI must move beyond static scoring to provide adaptive, real-time guidance based on reliable skill assessment.

AI is already in the mental health consulting room, whether clinicians invite it or not. OpenAI reports more than 800 million regular users of ChatGPT and more than 40 million people turning to the platform daily for health questions, yet the evidence remains limited on the clinical efficacy of large language model–based mental health chatbots. Professional guidance and regulation remain fragmented. Clinicians are therefore practicing in a gap between a patient reality that cannot be ignored and a professional infrastructure not yet built. We argue that neither enthusiastic adoption nor principled abstention is sufficient in the current environment. The clinician who refuses to discuss clinically material patient AI use is not preventing that use; they may lose visibility into a factor materially affecting the therapeutic process. The clinician who adopts AI without structured evaluation exposes patients and practice to avoidable harm. The stance we propose is structured harm reduction within the therapeutic relationship: inviting discussion at intake, integrating material use into case formulation, discussing benefits and limits, establishing crisis boundaries, monitoring throughout treatment, and addressing termination planning. Clinician-initiated use is a separate choice requiring evidence, consent when applicable, privacy safeguards, and accountability. We propose a phased clinical framework spanning preintake to termination anchored by the gather, understand, inform, document, and evaluate (GUIDE) mnemonic as an operational wrapper for daily practice. GUIDE and the phased framework are conceptual, unvalidated proposals. We distinguish ethical recommendations, established legal duties, and predicted standards; scope the legal discussion to the United States; integrate equity and subgroup performance; and offer proportionate implementation. The patients who seek care in an AI-saturated world deserve professional infrastructure capable of holding innovation and patient protection in informed balance.

AI systems are increasingly deployed across National Health Service (NHS) services, yet safety and implementation challenges may only become apparent after clinical go-live. Existing governance and implementation frameworks provide valuable high-level guidance, but health care provider organizations still require practical, auditable tools to support preimplementation decision-making.