Ensuring the safety and reliability of large language models (LLMs) in clinical practice is critical to prevent patient harm and promote trustworthy healthcare applications of AI. However, LLMs are advancing so rapidly that static safety benchmarks often become obsolete upon publication, yielding only an incomplete and sometimes misleading picture of model trustworthiness. We demonstrate that a Dynamic, Automatic, and Systematic (DAS) red-teaming framework that continuously stress-tests LLMs can reveal significant weaknesses of current LLMs across four safety-critical domains: robustness, privacy, bias/fairness, and hallucination. A suite of adversarial agents is applied to autonomously mutate test cases, identify/evolve unsafe-triggering strategies, and evaluate responses, uncovering vulnerabilities in real time without human intervention. Applying DAS to 15 proprietary and open-source LLMs revealed a stark contrast between static benchmark performance and vulnerability under adversarial pressure. Despite a median MedQA accuracy exceeding 80\%, 94\% of previously correct answers failed our dynamic robustness tests. We observed similarly high failure rates across other domains: privacy leaks were elicited in 86\% of scenarios, cognitive-bias priming altered clinical recommendations in 81\% of fairness tests, and we identified hallucination rates exceeding 66\% in widely used models. Such profound residual risks are incompatible with routine clinical practice. By converting red-teaming from a static checklist into a dynamic stress-test audit, DAS red-teaming offers the surveillance that hospitals/regulators/technology vendors require as LLMs become embedded in patient chatbots, decision-support dashboards, and broader healthcare workflows. Our framework delivers an evolvable, scalable, and reliable safeguard for the next generation of medical AI.

As artificial intelligence systems increasingly inform high-stakes decisions across sectors, transparency has become foundational to responsible and trustworthy AI implementation. Leveraging our role as a leading institute in advancing AI research and enabling industry adoption, we present key insights and lessons learned from practical interpretability applications across diverse domains. This paper offers actionable strategies and implementation guidance tailored to organizations at varying stages of AI maturity, emphasizing the integration of interpretability as a core design principle rather than a retrospective add-on.

Recent advances in generative models, particularly diffusion and auto-regressive models, have revolutionized fields like computer vision and natural language processing. However, their application to structure-based drug design (SBDD) remains limited due to critical data constraints. To address the limitation of training data for models targeting SBDD tasks, we propose an evolutionary framework named MEVO, which bridges the gap between billion-scale small molecule dataset and the scarce protein-ligand complex dataset, and effectively increase the abundance of training data for generative SBDD models. MEVO is composed of three key components: a high-fidelity VQ-VAE for molecule representation in latent space, a diffusion model for pharmacophore-guided molecule generation, and a pocket-aware evolutionary strategy for molecule optimization with physics-based scoring function. This framework efficiently generate high-affinity binders for various protein targets, validated with predicted binding affinities using free energy perturbation (FEP) methods. In addition, we showcase the capability of MEVO in designing potent inhibitors to KRAS$^{\textrm{G12D}}$, a challenging target in cancer therapeutics, with similar affinity to the known highly active inhibitor evaluated by FEP calculations. With high versatility and generalizability, MEVO offers an effective and data-efficient model for various tasks in structure-based ligand design.

Drug design, particularly in the discovery of lead compounds, is of core strategic importance to combating disease and enhancing human well-being. Prevailing computational methods, however, struggle to effectively integrate domain-specific knowledge, severely limiting their capacity to identify novel lead compounds with validated binding modes and new scaffolds. Here, we propose AutoLeadDesign, a lead compounds design framework that inspires extensive domain knowledge encoded in large language models with chemical fragments to progressively implement efficient exploration of vast chemical space. The comprehensive experiments indicate that AutoLeadDesign outperforms baseline methods. Significantly, empirical lead design campaigns targeting two clinically relevant targets (PRMT5 and SARS-CoV-2 PLpro) demonstrate AutoLeadDesign's competence in de novo generation of lead compounds, achieving expert-competitive design efficacy. Structural analysis further confirms their mechanism-validated inhibitory patterns. By tracing the process of design, we find that AutoLeadDesign shares analogous mechanisms with fragment-based drug design, which traditionally rely on expert decision-making, further revealing why it works. Overall, AutoLeadDesign offers an efficient approach for lead compound design, suggesting its potential utility in drug design.

Digital twins are models of real-world systems that can simulate their dynamics in response to potential actions. In complex settings, the state and action variables, and available data and knowledge relevant to a system can constantly change, requiring digital twins to continuously update with these changes to remain relevant. Current approaches struggle in this regard, as they require fixed, well-defined modelling environments, and they cannot adapt to novel variables without re-designs, or incorporate new information without re-training. To address this, we frame digital twinning as an in-context learning problem using large language models, enabling seamless updates to the twin at inference time. We develop CALM-DT, a Context-Adaptive Language Model-based Digital Twin that can accurately simulate across diverse state-action spaces using in-context learning alone by utilising fine-tuned encoders for sample retrieval. We empirically demonstrate CALM-DT's competitive performance with existing digital twin approaches, and its unique ability to adapt to changes in its modelling environment without parameter updates.

Madeline Gardner is the youngest American person to get a Hero Arm, the world's first multi-grip bionic arm for children. Kicking a ball or climbing stairs with ease after a leg amputation above the knee is now within reach. Researchers at MIT have developed a bionic knee that redefines mobility for above-the-knee amputees. The result is faster, smoother, and more natural movement. Led by Professor Hugh Herr, the MIT team created a solution that doesn't just mimic motion, it responds to intent.

An Apple Watch saved his life after it used SOS to call for help when he had a stroke in his driveway. What if your Apple Watch or iPhone could alert you to a pregnancy before a test does? A new Apple-funded study suggests that this is now within reach. Researchers used a mix of behavioral and biometric data to train an artificial intelligence model. The system correctly predicted pregnancy in 92% of cases.

Traditional Chinese Medicine diagnosis and treatment principles, established through centuries of trial-and-error clinical practice, directly maps patient-specific symptom patterns to personalised herbal therapies. These empirical holistic mapping principles offer valuable strategies to address remaining challenges of reductionism methodologies in modern biomedicine. However, the lack of a quantitative framework and molecular-level evidence has limited their interpretability and reliability. Here, we present an AI framework trained on ancient and classical TCM formula records to quantify the symptom pattern-herbal therapy mappings. Interestingly, we find that empirical TCM diagnosis and treatment are consistent with the encoding-decoding processes in the AI model. This enables us to construct an interpretable TCM embedding space (TCM-ES) using the model's quantitative representation of TCM principles. Validated through broad and extensive TCM patient data, the TCM-ES offers universal quantification of the TCM practice and therapeutic efficacy. We further map biomedical entities into the TCM-ES through correspondence alignment. We find that the principal directions of the TCM-ES are significantly associated with key biological functions (such as metabolism, immune, and homeostasis), and that the disease and herb embedding proximity aligns with their genetic relationships in the human protein interactome, which demonstrate the biological significance of TCM principles. Moreover, the TCM-ES uncovers latent disease relationships, and provides alternative metric to assess clinical efficacy for modern disease-drug pairs. Finally, we construct a comprehensive and integrative TCM knowledge graph, which predicts potential associations between diseases and targets, drugs, herbal compounds, and herbal therapies, providing TCM-informed opportunities for disease analysis and drug development.

Individuals are increasingly relying on large language model (LLM)-enabled conversational agents for emotional support. While prior research has examined privacy and security issues in chatbots specifically designed for mental health purposes, these chatbots are overwhelmingly "rule-based" offerings that do not leverage generative AI. Little empirical research currently measures users' privacy and security concerns, attitudes, and expectations when using general-purpose LLM-enabled chatbots to manage and improve mental health. Through 21 semi-structured interviews with U.S. participants, we identified critical misconceptions and a general lack of risk awareness. Participants conflated the human-like empathy exhibited by LLMs with human-like accountability and mistakenly believed that their interactions with these chatbots were safeguarded by the same regulations (e.g., HIPAA) as disclosures with a licensed therapist. We introduce the concept of "intangible vulnerability," where emotional or psychological disclosures are undervalued compared to more tangible forms of information (e.g., financial or location-based data). To address this, we propose recommendations to safeguard user mental health disclosures with general-purpose LLM-enabled chatbots more effectively.

Recent advances in protein structure prediction have achieved near-atomic accuracy for well-folded proteins. However, current benchmarks inadequately assess model performance in biologically challenging contexts, especially those involving intrinsically disordered regions (IDRs), limiting their utility in applications such as drug discovery, disease variant interpretation, and protein interface design. We introduce DisProtBench, a comprehensive benchmark for evaluating protein structure prediction models (PSPMs) under structural disorder and complex biological conditions. DisProtBench spans three key axes: (1) Data complexity, covering disordered regions, G protein-coupled receptor (GPCR) ligand pairs, and multimeric complexes; (2) Task diversity, benchmarking twelve leading PSPMs across structure-based tasks with unified classification, regression, and interface metrics; and (3) Interpretability, via the DisProtBench Portal, which provides precomputed 3D structures and visual error analyses. Our results reveal significant variability in model robustness under disorder, with low-confidence regions linked to functional prediction failures. Notably, global accuracy metrics often fail to predict task performance in disordered settings, emphasizing the need for function-aware evaluation. DisProtBench establishes a reproducible, extensible, and biologically grounded framework for assessing next-generation PSPMs in realistic biomedical scenarios.