Headlines

Dr. Donald Kohn has been developing gene therapies for rare pediatric immune disorders for over 30 years. This week, his role in a clinical trial has culminated in the first-ever U.S. Food and Drug Administration–approved therapy for severe leukocyte adhesion deficiency-I—a genetic condition characterized by recurrent infections and, often, early death.

Researchers are exploring the prospect of using gut bacteria to boost muscle strength, after zeroing in on a microbe that does this in mice

Dr. Donald Kohn has been developing gene therapies for rare pediatric immune disorders for over 30 years.

In the months before his wedding, Queens resident Roberto Pineda Sanchez noticed persistent rectal bleeding, a dangerous symptom he almost ignored.

Dynamic switching between revival stem cells and conventional intestinal stem cells enables efficient tissue repair without exhausting the stem cell pool, report researchers from Institute of Science Tokyo. Using organoid and mouse disease models, the researchers uncovered how flexible stress-tolerant cell states contribute to intestinal repair—providing a better understanding of the biological mechanisms driving intestinal regeneration.

AI-enabled medical devices promise improved medical care and support for health care professionals. However, the safety and performance of such systems not only depends on algorithms or technical specifications. It is equally important how people use these devices and applications. In a recent publication in the journal NEJM AI, a research team led by Prof. Stephen Gilbert from Else Kröner Fresenius Center (EKFZ) for Digital Health at TUD Dresden University of Technology systematically analyzes risks that can arise in human-AI interactions and makes recommendations for manufacturers and regulatory evaluators.

SiliconRepublic.com spoke with experts at Amgen to explore how early career guidance can set the foundations for a happy and productive career. Read more: How does mentorship impact a professional’s career longevity?

Marne-cel now offers an option to children with the rare immune disorder who lack a sibling donor for stem cell transplantation.

A rare disease gene therapy from Rocket Pharmaceuticals has garnered FDA approval after an earlier rejection for manufacturing problems. The FDA on Thursday granted accelerated approval to Rocket Pharma’s gene therapy ...

This GEN Live show will focus on new areas of stem cell research, including aging. It will bring together a panel of leading experts to unpack the latest advances and ongoing challenges in stem cells. The post Stem Cells at the Frontier: Breakthroughs, Barriers, and What’s Next appeared first on GEN - Genetic Engineering and Biotechnology News.

Artificial intelligence (AI) could help physicians determine if survivors of childhood cancer need extra support—and the more information included in AI prompting, the better its performance. This finding, published in Communications Medicine by scientists from St. Jude Children's Research Hospital, may guide future integration of AI into clinical workflows.

A US jury has ruled that deliberately addictive features of apps owned by Meta and Google harmed a young user’s mental health.The ruling, by the California Superior Court of Los Angeles, found that the platforms were deliberately designed in ways that contributed to the anxiety, depression, and body dysmorphia of the plaintiff KGM, as she is referred to in court documents, now a 20 year old woman.KGM’s lawyers argued that algorithmic recommendations, “likes,” push notifications, and autoplaying videos, among other features, were engineered to encourage compulsive use and exploited developmental vulnerabilities in KGM. This, they claimed, had contributed to a deterioration in her mental health.Meta reportedly argued that KGM’s mental health problems, including body dysmorphia and thoughts of self-harm, were caused by familial abuse and turmoil.1 However, the company added that it no longer set internal targets for user time spent on its platform.The video sharing platform YouTube, owned

Discover the future of pet safety with SATELLAI. From 5-way GNSS satellite tracking to PetSense AI health analysis, these smart collars offer precision and peace of mind for every dog and terrain.

arXiv:2603.24986v1 Announce Type: new Abstract: Large language model based health agents are increasingly used by health consumers and clinicians to interpret health information and guide health decisions. However, most AI systems in healthcare operate in siloed configurations, supporting individual users rather than the multi-stakeholder relationships central to healthcare. Such use can fragment understanding and exacerbate misalignment among patients, caregivers, and clinicians. We reframe AI not as a standalone assistant, but as a collaborator embedded within multi-party care interactions. Through a clinically validated fictional pediatric chronic kidney disease case study, we show that breakdowns in adherence stem from fragmented situational awareness and misaligned goals, and that siloed use of general-purpose AI tools does little to address these collaboration gaps. We propose a conceptual framework for designing AI collaborators that surface contextual information, reconcile

Japanese news outlet Akiba PC Hotline! examines a fake Samsung 990 Pro SSD that's starting to show up in the Japanese market.

This week I reported on some rather unusual research that focuses on the brain of L. Stephen Coles. Coles was a gerontologist who died from pancreatic cancer in 2014. He had spent the latter part of his career specializing in human longevity. And before he died, he decided to have his brain preserved by a…

Lily Mae Lazarus / Fortune: NY-based Blossom Health, which makes an “AI copilot” to augment psychiatrists' clinical decisions and automate office tasks, raised $20M in seed and Series A  —  Blossom Health has raised $20 million in seed and Series A funding to bring an AI “copilot” for psychiatry to patients nationwide, Fortune has exclusively learned.

Despite an urgent demand for kidney donors in the U.S., about 1 in 4 donor kidneys goes unused. In 2022, more than 71,000 people were on the kidney transplant waiting list, with waitlists often extending as long as five years for a donor kidney.

CRISPR is a powerful DNA-editing tool that has underpinned huge advancements in human health care in the last decade. It is a precision tool, but is not perfect, and misplaced DNA edits can compromise safety and efficacy, costing billions each year. Researchers at the MRC Laboratory of Medical Sciences (LMS), Imperial College London and the University of Sheffield have published research in Nature showing that the physical twisting of DNA plays an important role in these mistakes. Using a newly developed platform of tiny (nanometer-sized) DNA circles, called DNA minicircles, the team captured never-before-seen interactions between CRISPR and DNA, providing insights that could help eradicate errors altogether.

Structural basis of supercoiling-induced CRISPR–Cas9 off-target activity  Nature

Structural basis of supercoiling-induced CRISPR–Cas9 off-target activity  Nature

Blossom Health, a New York-based telepsychiatry startup founded in 2024, has raised $20 million in combined seed and Series A funding to scale an AI-powered platform that pairs psychiatrists with clinical copilots and automated administrative support. The round was led by Headline, whose co-founder and managing partner Mathias Schilling is joining the company’s board. Village […] This story continues at The Next Web

A few qualities, including a sense of purpose, seem to have real benefits — especially as you age.

Rasmus Dahlin's fiancée Carolina Matovac returned to a Sabres game months after suffering major heart failure and receiving a transplant in France.

Tissues harbor memories of inflammation, which heighten sensitivity to diverse future assaults. Whether and how these adaptations are sustained through time and cell division remain poorly understood. We show that in mice, epidermal stem cells store ...

While Minnesota tries to make sure cell-cultured food is labeled as such, South Dakota has passed a moratorium on any lab-grown meat in the state.

Hoth Therapeutics launches OpenClaw AI platform to speed drug discovery, improve data integration, and boost development efficiency. Importance Rank:  1 read more

How a Healthy Mind-Set Influences Longevity  The New York TimesPeople 65 and older can get better with age, study shows. This is the key.  The Washington PostThe condition of many elderly people actually improves over the years – and this is the reason  The Jerusalem PostA 103-Year-Old Woman Says This Simple Habit Is The Secret To A Long Life — And Research Backs Her Up  YourTangoAgeing does not always mean decline  diabetes.co.uk

How a Healthy Mind-Set Influences Longevity  The New York TimesPeople 65 and older can get better with age, study shows. This is the key.  The Washington PostThe condition of many elderly people actually improves over the years – and this is the reason  The Jerusalem PostA 103-Year-Old Woman Says This Simple Habit Is The Secret To A Long Life — And Research Backs Her Up  YourTangoAgeing does not always mean decline  diabetes.co.uk

arXiv:2603.24380v1 Announce Type: new Abstract: We study higher-order curvature estimates along K\"ahler-Ricci flows on compact K\"ahler manifolds of intermediate Kodaira dimension. We prove that away from singular fibers, the Ricci curvature is uniformly bounded in $C^1$, the Laplacian of the Ricci curvature in $C^0$, and the scalar curvature in $C^2$. We identify a geometric obstruction to higher-order curvature bounds, whose non-vanishing causes a specific third-order derivative of the Ricci curvature to blow up at rate $e^{t/2}$. Uniform $C^k$ bounds for every $k$ hold for the Ricci curvature in the isotrivial case, and for the full Riemann curvature in the torus-fibered case.

arXiv:2603.23890v1 Announce Type: new Abstract: As the modern microservice architecture for cloud applications grows in popularity, cloud services are becoming increasingly complex and more vulnerable to misconfiguration and software bugs. Traditional approaches rely on expert input to diagnose and fix microservice anomalies, which lacks scalability in the face of the continuous integration and continuous deployment (CI/CD) paradigm. Microservice rollouts, containing new software installations, have complex interactions with the components of an application. Consequently, this added difficulty in attributing anomalous behavior to any specific installation or rollout results in potentially slower resolution times. To address the gaps in current diagnostic methods, this paper introduces Praxium, a framework for anomaly detection and root cause inference. Praxium aids administrators in evaluating target metric performance in the context of dependency installation information provided by

Peoples’ use of AI for researching health information and supporting mental health appears to trump consumers’ trust in the technology, based on the March 2026 Kaiser Family Foundation (KFF) Tracking Poll on Health Information and Trust.          Similar to the recent Rock Health consumer survey on AI adoption (discussed in Health Populi earlier this week), about 1 in 3 U.S. adults have used AI for health information in the past year for either physical or mental health queries.              There are many studies gauging consumers’ use of AI in health care, but KFF uniquely surfaced an important nuance in this study regarding just “who” is more likely to use AI for health: that is that younger and lower income adults were much more likely to cite not being able to access or afford health care, and were thus motivated to turn to AI for health advice.                

Cathie Wood's Ark Invest bought Tempus AI shares while trimming Meta, signaling a shift toward medical AI over big tech exposure. Importance Rank:  1 read more

Click here for more articles by Kormedi.com. A new study suggests that what you eat may influence proteins inside your cells, specifically those in the mitochondria. The Mediterranean diet — rich in olive oil, vegetables, fruits, whole grains, fish and legumes, while limiting red meat and processed foods — is already known to benefit heart and brain health. However, exactly how it works in the body has remained unclear. Study sheds light on what’s happening inside cells Researchers from the University of Southern California and Sapienza University of Rome found that people who closely followed the Mediterranean diet had higher levels of two mitochondrial proteins. The study involved 49 elderly patients (average age of 78) with non-valvular atrial fibrillation, a condition that increases stroke risk. These patients tend to have high oxidative stress — a condition where harmful reactive oxygen species outweigh the body’s defenses — making it easier to detect diet-related changes.

In two different cases on Wednesday, United States authorities indicted Chinese nationals on charges ranging from conspiring to smuggle advanced AI chips to China to drug trafficking and money laundering. The cases came on the same day US President Donald Trump announced new mid-May dates for his highly anticipated summit with China. The developments come as the two countries continue to compete over global leadership in artificial intelligence, and as Washington continues to accuse Beijing of...

A research team led by Purdue University's W. Andy Tao has discovered a new type of protein modification related to cellular mutation that impairs a crucial enzyme's ability to help drive energy processes. Their discovery, published in Nature Chemistry, opens a new route to therapeutic cancer intervention.

If responsibly developed and thoughtfully integrated into healthcare systems, regenerative therapies could help shift medicine toward a more proactive model, i.e., one focused on preserving health, maintaining function, and extending healthy lifespan. The post Leveraging the Full Potential of Regenerative Medicine Requires a Proactive Approach appeared first on GEN - Genetic Engineering and Biotechnology News.

In back-to-back studies published in Nature, researchers from Purdue University and Columbia University report a naturally evolved gene-editing system that can activate genes, offering an advantage over existing CRISPR gene-editing systems that merely find and cut DNA. The research includes two complementary studies, one examining the biological function of the system and the other revealing the molecular mechanism that enables it.

Researchers have identified many contributing issues leading to the characteristic loss of muscle mass and strength that takes place with age. Arguably the central problems are (a) the disruptions of cell behavior caused by chronic inflammation, (b) damage to neuromuscular junctions, depriving muscle tissue of signals it relies upon for normal maintenance to take place, and (c) loss of muscle stem cell activity, and thus a reduced supply of somatic muscle cells to replace losses. These central problems likely interact with one another, but in principle could be addressed distinctly to produce benefits in patients. Past studies have shown, rather convincingly, that muscle stem cells in older individuals retain their function when moved from an old environment to a young environment. The problem is not […]

The same principles that help astronauts stay strong in microgravity can help us all resist the slow collapse of ageing – and it’s not all about hitting the gym more

A 20-year study has shown that, like photocopying photocopies, cloning doesn't produce perfect copies – with big implications for farming, conservation and de-extinction

The latest news and headlines from Yahoo! News. Get breaking news stories and in-depth coverage with videos and photos.

BullFrog AI launches bfARENAS, a scenario-based platform to improve R&D decisions and reduce risk in drug development portfolios. Importance Rank:  1 read more

Longevity medicine and anti aging clinics promise longer, healthier lives with biomarker testing, gene therapy, and personalized treatments.

It has been fifteen years since the first compelling demonstration of clearance of senescent cells in mice. That study paved the way for the transformation of the research community into one convinced of the relevance of cellular senescence to degenerative aging. It also helped to change the culture of aging research more generally, one of the important contributions to a shift in attitudes that has led to a research and development community that understands the treatment of aging as a medical condition to be a practical, desirable goal. Here, discuss the role of cellular senescence in muscle aging specifically; how it contributes to harm and lost function, and what might be done about it. Cellular senescence is increasingly recognized as a pivotal mechanism driving skeletal […]

While gastroenterology patients generally support AI as an assistive tool, they prioritize physician oversight and transparency.

Building your butt muscles will help you stay injury free and independent in midlife and beyond.

arXiv:2603.22344v1 Announce Type: cross Abstract: Large language models (LLMs) assisted literature retrieval may lead to erroneous references, but these errors have not been rigorously quantified. Therefore, we quantitatively assess errors in reference retrieval of widely used free-version LLM platforms and identify the factors associated with retrieval errors. We evaluated 2,000 references retrieved by 5 LLMs (Grok-2, ChatGPT GPT-4.1, Google Gemini Flash 2.5, Perplexity AI, and DeepSeek GPT-4) for 40 randomly-selected original articles (10 per journal) published Jan. 2024 to July 2025 from British Medical Journal (BMJ), Journal of the American Medical Association, and The New England Journal of Medicine (NEJM). Primary outcomes were a multimetric score ratio combining validity of digital object identifier, PubMed ID, Google-Scholar link, and relevance; and complete miss rate (proportion of references failing all applicable metrics). Multivariable regression was used to examine

arXiv:2603.23358v1 Announce Type: new Abstract: This proof-of-concept study introduces a novel multimodal framework combining synchronized EEG-fNIRS modalities with neuronal avalanche analysis to identify early network dysfunction in Alzheimer's disease. The approach leverages complementary neural signals to examine motor network dynamics during execution and imagery tasks within an interactive task environment. Preliminary analysis of a small pilot cohort (N=4 subjects, including one with Mild Cognitive Impairment) validated the technical feasibility of the multimodal framework and revealed observable condition-dependent patterns in network organization. Two primary observations emerged: a reduced neural contrast between motor execution and imagery states, and increased trial-to-trial variability in network organization in the MCI participant. These initial results successfully validate the technical pipeline and provide hypothesis-generating observations for future statistically

arXiv:2603.22731v1 Announce Type: new Abstract: Autonomous mobile robot fleets must coordinate task allocation and charging under limited shared resources, yet most battery aware planning methods address only a single robot. This paper extends degradation cost aware task planning to a multi robot setting by jointly optimizing task assignment, service sequencing, optional charging decisions, charging mode selection, and charger access while balancing degradation across the fleet. The formulation relies on reduced form degradation proxies grounded in the empirical battery aging literature, capturing both charging mode dependent wear and idle state of charge dependent aging; the bilinear idle aging term is linearized through a disaggregated piecewise McCormick formulation. Tight big M values derived from instance data strengthen the LP relaxation. To manage scalability, we propose a hierarchical matheuristic in which a fleet level master problem coordinates assignments, routes, and

arXiv:2603.22727v1 Announce Type: new Abstract: This work proposes a novel immersive communication framework that leverages brain-computer interface (BCI) to acquire brain signals for inferring user-centric states (e.g., intention and perception-related discomfort), thereby enabling more personalized and robust immersive adaptation under strong individual variability. Specifically, we develop a personalized federated learning (PFL) model to analyze and process the collected brain signals, which not only accommodates neurodiverse brain-signal data but also prevents the leakage of sensitive brain-signal information. To address the energy bottleneck of continual on-device learning and inference on energy-limited immersive terminals (e.g., head-mounted display), we further embed spiking neural networks (SNNs) into the PFL. By exploiting sparse, event-driven spike computation, the SNN-enabled PFL reduces the computation and energy cost of training and inference while maintaining

arXiv:2603.22447v1 Announce Type: new Abstract: Agent skills are modular instruction packages that combine YAML metadata, natural language instructions, and embedded code, and they have reached 196K publicly available instances, yet no mechanism exists to detect clone relationships among them. This gap creates systemic risks: a vulnerability in a widely copied skill silently persists across derivatives with no alert to maintainers. Existing clone detectors, designed for single-modality source code, cannot handle the multi-modal structure of skills, where clone evidence is distributed across three interleaved content channels. We present SkillClone, the first multi-modal clone detection approach for agent skills. SkillClone fuses flat TF-IDF similarity with per-channel decomposition (YAML, NL, code) through logistic regression, combining strong detection with interpretable type classification. We construct SkillClone-Bench, a balanced benchmark of 300 ground-truth pairs with stratified

arXiv:2603.22344v1 Announce Type: new Abstract: Large language models (LLMs) assisted literature retrieval may lead to erroneous references, but these errors have not been rigorously quantified. Therefore, we quantitatively assess errors in reference retrieval of widely used free-version LLM platforms and identify the factors associated with retrieval errors. We evaluated 2,000 references retrieved by 5 LLMs (Grok-2, ChatGPT GPT-4.1, Google Gemini Flash 2.5, Perplexity AI, and DeepSeek GPT-4) for 40 randomly-selected original articles (10 per journal) published Jan. 2024 to July 2025 from British Medical Journal (BMJ), Journal of the American Medical Association, and The New England Journal of Medicine (NEJM). Primary outcomes were a multimetric score ratio combining validity of digital object identifier, PubMed ID, Google-Scholar link, and relevance; and complete miss rate (proportion of references failing all applicable metrics). Multivariable regression was used to examine

arXiv:2603.22322v1 Announce Type: new Abstract: Machine learning systems deployed in medical devices require governance frameworks that ensure safety while enabling continuous improvement. Regulatory bodies including the FDA and European Union have introduced mechanisms such as the Predetermined Change Control Plan (PCCP) and Post-Market Surveillance (PMS) to manage iterative model updates without repeated submissions. This paper presents AI/ML Evaluation and Governance Infrastructure for Safety (AEGIS), a governance framework applicable to any healthcare AI system. AEGIS comprises three modules, i.e., dataset assimilation and retraining, model monitoring, and conditional decision, that operationalize FDA PCCP and EU AI Act Article 43(4) provisions. We implement a four-category deployment decision taxonomy (APPROVE, CONDITIONAL APPROVAL, CLINICAL REVIEW, REJECT) with an independent PMS ALARM signal, enabling detection of the critical state in which no deployable model exists while the

At Yahoo Finance, you get free stock quotes, up-to-date news, portfolio management resources, international market data, social interaction and mortgage rates that help you manage your financial life.

Technological innovation has always driven surgical progress, and AI now represents the next transformative wave. Machine learning models are now being developed to predict surgical risks, assist in diagnosing rare congenital disorders, analyze imaging data, and anticipate postoperative complications. Risk prediction tools have already shifted from traditional statistical methods to more complex machine learning approaches, improving their ability to account for nonlinear interactions.

The UK has failed to keep pace with the rest of the world. Can it regain its status, and how?

Статья на конкурс «Био/Мол/Текст»: В данной статье мы разберем одну из самых неоднозначных тем в медицине и науке — longevity (далее в статье мы поясним, что это такое), постараемся систематизировать популярные высказывания ученых, врачей и простых энтузиастов, чтобы создать общую картину развития движения борьбы со старением и помочь вам сформировать свое мнение насчет «лечения» старения, а не слепо верить словам «экспертов».

Microscopic sensors that are as thin as a strand of hair but capable of taking multiple measurements simultaneously could revolutionize the diagnosis and monitoring of diseases like cancer.

'Dead End': Radical 20-Year Study Reveals Genetic Cloning Hits a Limit  ScienceAlertMouse study shows repeated cloning causes grave genetic mutations  ReutersLimitations of serial cloning in mammals  NatureScientists Tried to Clone Clones Forever. It Didn’t End Well  GizmodoMammals cannot be cloned infinitely, mice study discovers  France 24

'Dead End': Radical 20-Year Study Reveals Genetic Cloning Hits a Limit  ScienceAlertMouse study shows repeated cloning causes grave genetic mutations  ReutersLimitations of serial cloning in mammals  NatureScientists Tried to Clone Clones Forever. It Didn’t End Well  GizmodoMammals cannot be cloned infinitely, mice study discovers  France 24

Artificial intelligence holds promise for helping doctors diagnose patients and personalize treatment options. However, an international group of scientists led by MIT cautions that AI systems, as currently designed, carry the risk of steering doctors in the wrong direction because they may overconfidently make incorrect decisions.

Scientists Tried to Clone Clones Forever. It Didn’t End Well  GizmodoMouse study shows repeated cloning causes grave genetic mutations  ReutersLimitations of serial cloning in mammals  Nature'Dead End': Radical 20-Year Study Reveals Genetic Cloning Hits a Limit  ScienceAlertMammals cannot be cloned infinitely, mice study discovers  France 24

Scientists Tried to Clone Clones Forever. It Didn’t End Well  GizmodoMouse study shows repeated cloning causes grave genetic mutations  ReutersLimitations of serial cloning in mammals  Nature'Dead End': Radical 20-Year Study Reveals Genetic Cloning Hits a Limit  ScienceAlertMammals cannot be cloned infinitely, mice study discovers  France 24

A new research paper was published in Volume 17 of Oncotarget on March 17, 2026, titled "CREB5 regulates stem cell-like transcriptional programs to enhance tumor progression in prostate cancer."

There is a limit on how many times a mammal can be cloned before suffering "mutational meltdown," Japanese scientists have discovered, after making 1,200 clones over two decades that started off with a single mouse.

The latest news and headlines from Yahoo! News. Get breaking news stories and in-depth coverage with videos and photos.

Ocugen said it will take its gene therapy for a form of vision loss into a Phase 3 trial, after reporting mixed results on Tuesday from a smaller study. The gene therapy is for patients ...

In public discourse, the increasing lifespan in Western countries is often linked to longer life in good health. However, studying human aging in modern societies is complex because outcomes are shaped by numerous social, behavioral, and environmental factors, including medical advances, food security, poverty, alcohol use, and civil violence.

TUESDAY, March 24, 2026 — Folks who rely on chatbots for their scientific and medical info, be forewarned — artificial intelligence (AI) gets a "D" when it’s asked to evaluate whether a claim is true or false, a new study says.ChatGPT’s accuracy in...

AI-generated X-rays are now so realistic they could fool doctors—and potentially disrupt the entire healthcare system. A new study published today (March 24) in Radiology, the journal of the Radiological Society of North America (RSNA), finds that both radiologists and advanced multimodal large language models (LLMs) struggle to reliably tell apart real X-rays from artificial [...]

The Vatican said on Tuesday that Catholics can receive transplants ⁠of animal tissues to address medical conditions, as procedures involving genetically modified pig or cow...

Insilico Medicine (3696.HK), a clinical-stage drug discovery and development company driven by generative artificial intelligence (AI), today announced a strategic research collaboration with ASKA Pharmaceutical Co., Ltd. ("ASKA"), a specialized pharmaceutical company with a strong focus on internal medicine, obstetrics, and gynecology.

Triangle Health raised $4 million to help people understand their health conditions and explore treatment options, Endpoints News learned exclusively. After a person uploads their health records, Triangle’s AI can brainstorm and compare treatments, prepare ...

The latest news and headlines from Yahoo! News. Get breaking news stories and in-depth coverage with videos and photos.

A global team of leading scientists is zeroing in on a tiny but powerful molecule that could reshape how we age. Known as NAD⁺, it plays a crucial role in keeping our cells energized, repairing DNA, and maintaining overall health—but its levels steadily decline over time, potentially fueling diseases like Alzheimer’s and Parkinson’s. Researchers are now exploring ways to boost NAD⁺ using compounds like NR and NMN, with early studies hinting at improvements in memory, metabolism, and physical function.

From pre- and self-diagnosis of symptoms to prescription drug treatments and ongoing care, millions of U.S. consumers have used AI chatbots for health and wellness according to the consumer health adoption survey from Rock Health. The report was published in late March 2026, with AI-focused data from the consumer survey fielded in December 2025.              In just one year from 2024 to 2025, consumers’ use of AI chatbots for health information doubled from 16% of people to 32%. The most popular brand of AI for health information seeking was ChatGPT (owned by OpenAI, with whom Microsoft is a major partner), used by 23% of seekers, followed by Gemini (from Google) for 15% of consumers using AI for health. Other vendors of chatbots for health garnered lower single-digits in share of users.            As we might anticipate, healthcare AI users tend to be digital health tech consumers for other devices and apps more than non-AI

arXiv:2603.20538v1 Announce Type: cross Abstract: Behavior cloning is a fundamental paradigm in machine learning, enabling policy learning from expert demonstrations across robotics, autonomous driving, and generative models. Autoregressive models like transformer have proven remarkably effective, from large language models (LLMs) to vision-language-action systems (VLAs). However, applying autoregressive models to continuous control requires discretizing actions through quantization, a practice widely adopted yet poorly understood theoretically. This paper provides theoretical foundations for this practice. We analyze how quantization error propagates along the horizon and interacts with statistical sample complexity. We show that behavior cloning with quantized actions and log-loss achieves optimal sample complexity, matching existing lower bounds, and incurs only polynomial horizon dependence on quantization error, provided the dynamics are stable and the policy satisfies a

arXiv:2603.22106v1 Announce Type: new Abstract: Over time, the shared understanding that makes a software system safe to change quietly erodes. This gradual loss of understanding across a team increases cognitive debt, while the loss of captured rationale leads to intent debt. These may become more important, than technical debt in AI-assisted software development. This article proposes a triple debt model to reason about software health. It is built around three interacting debt types: technical debt in code, cognitive debt in people, and intent debt in externalized knowledge. Cognitive debt concerns what people understand; intent debt concerns what is explicitly captured for both people and machines to use in the future.

arXiv:2603.21656v1 Announce Type: new Abstract: Protecting patient privacy remains a fundamental barrier to scaling machine learning across healthcare institutions, where centralizing sensitive data is often infeasible due to ethical, legal, and regulatory constraints. Federated learning offers a promising alternative by enabling privacy-preserving, multi-institutional training without sharing raw patient data; however, real-world deployments face severe challenges from data heterogeneity, site-specific biases, and class imbalance, which degrade predictive reliability and render existing uncertainty quantification methods ineffective. Here, we present TrustFed, a federated uncertainty quantification framework that provides distribution-free, finite-sample coverage guarantees under heterogeneous and imbalanced healthcare data, without requiring centralized access. TrustFed introduces a representation-aware client assignment mechanism that leverages internal model representations to

arXiv:2603.21460v1 Announce Type: new Abstract: Patient education materials for solid-organ transplantation vary substantially across U.S. centers, yet no systematic method exists to quantify this heterogeneity at scale. We introduce a framework that grounds the same patient questions in different centers' handbooks using retrieval-augmented language models and compares the resulting answers using a five-label consistency taxonomy. Applied to 102 handbooks from 23 centers and 1,115 benchmark questions, the framework quantifies heterogeneity across four dimensions: question, topic, organ, and center. We find that 20.8% of non-absent pairwise comparisons exhibit clinically meaningful divergence, concentrated in condition monitoring and lifestyle topics. Coverage gaps are even more prominent: 96.2% of question-handbook pairs miss relevant content, with reproductive health at 95.1% absence. Center-level divergence profiles are stable and interpretable, where heterogeneity reflects

arXiv:2603.20815v1 Announce Type: new Abstract: The pharmaceutical industry is facing challenges with quality management such as high costs of compliance, slow responses and disjointed knowledge. This paper presents GMPilot, a domain-specific AI agent that is designed to support FDA cGMP compliance. GMPilot is based on a curated knowledge base of regulations and historical inspection observations and uses Retrieval-Augmented Generation (RAG) and Reasoning-Acting (ReAct) frameworks to provide real-time and traceable decision support to the quality professionals. In a simulated inspection scenario, GMPilot shows how it can improve the responsiveness and professionalism of quality professionals by providing structured knowledge retrieval and verifiable regulatory and case-based support. Although GMPilot lacks in the aspect of regulatory scope and model interpretability, it is a viable avenue of improving quality management decision-making in the pharmaceutical sector using intelligent

arXiv:2603.20538v1 Announce Type: new Abstract: Behavior cloning is a fundamental paradigm in machine learning, enabling policy learning from expert demonstrations across robotics, autonomous driving, and generative models. Autoregressive models like transformer have proven remarkably effective, from large language models (LLMs) to vision-language-action systems (VLAs). However, applying autoregressive models to continuous control requires discretizing actions through quantization, a practice widely adopted yet poorly understood theoretically. This paper provides theoretical foundations for this practice. We analyze how quantization error propagates along the horizon and interacts with statistical sample complexity. We show that behavior cloning with quantized actions and log-loss achieves optimal sample complexity, matching existing lower bounds, and incurs only polynomial horizon dependence on quantization error, provided the dynamics are stable and the policy satisfies a

arXiv:2603.20441v1 Announce Type: new Abstract: Verification-guided self-improvement has recently emerged as a promising approach to improving the accuracy of large language model (LLM) outputs. However, existing approaches face a trade-off between inference efficiency and accuracy: iterative verification-rectification is computationally expensive and prone to being trapped in faulty reasoning, while best-of-N selection requires extensive sampling without addressing internal model flaws. We propose a training-free regeneration paradigm that leverages an offline-curated contrastive Reflection Memory (RM) to provide corrective guidance, while regenerating from scratch helps break out of faulty reasoning. At inference time, the method performs RM-guided self-verification followed by a single RM-guided regeneration, avoiding both iterative correction and multi-sample selection. We evaluated our method on nine benchmarks that span algorithmic, reasoning, symbolic, and domain-specific tasks

A new study led by researchers at the USC Leonard Davis School of Gerontology suggests that the benefits of the Mediterranean diet may be driven, in part, by tiny proteins hidden within our mitochondria, opening a new window into how diet shapes aging and disease risk.

A study led by researchers at the USC Leonard Davis School of Gerontology suggests that the benefits of the Mediterranean diet may be driven, in part, by tiny proteins hidden within our mitochondria, opening a new window into how diet shapes aging and disease risk. The findings are published in the journal Frontiers in Nutrition.

Chinese researchers have developed a novel and highly efficient mitochondrial capsule transplantation therapy, achieving the safe and efficient transplantation of healthy mitochondria into cells and tissues for the first time. This new therapy can significantly alleviate symptoms of severe diseases such as Parkinson's disease.

When stem cells are injected into thick tissues, they often clump together and die from lack of oxygen or nutrients, making it almost impossible to use them for thick, highly vascularized tissues like organ transplants or muscle repair.