Research on artificial intelligence (AI) and mental health has focused largely on harms at deployment, including chatbot safety, sycophancy, and AI-associated delusions. Less attention has been paid to a prior question: whether the human-generated text and preference judgments that shape large language models are themselves clinically reliable, particularly when self-report may be distorted. This Viewpoint aims to develop the clinical psychiatric construct of collusion—the uncritical acceptance of an unreliable account—as an analytic lens for AI training and deployment, and to argue that the clinical reliability of training and preference data should be treated as an explicit trustworthy-AI criterion in mental-health–relevant systems. A conceptual synthesis of psychiatry, clinical psychology, and AI safety literature was undertaken. The analysis distinguishes three pipeline layers: pretraining corpora, preference data and posttraining methods, and deployment-time interaction. It maps the clinical construct of collusion against adjacent technical concepts, including sycophancy, reward overoptimization, grounding, refusal training, red-teaming, and live monitoring. The synthesis suggests that collusion-like dynamics are least applicable at the pretraining layer and most applicable at the preference-data and deployment layers, where unassessed user or labeler input can be reinforced without corroboration. Existing mitigations, including data curation, Constitutional AI, reward-model evaluation, grounded generation, refusal training, red-teaming, and postdeployment monitoring, address parts of this problem. However, these approaches are not yet organized around a clinically informed account of when self-report is unreliable. The central novelty is therefore not a generic claim about bias, but the proposal that clinical self-report reliability should be assessed as a distinct data-quality and governance dimension. Trustworthy-AI frameworks for mental-health–relevant applications should incorporate clinical expertise in self-report reliability into preference-data design, red-teaming, and postmarket surveillance. Adding the clinical reliability of training and preference data as an explicit criterion could complement existing technical safeguards while leaving empirical evaluation of clinician involvement as an open research agenda.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/5de93d37e885a1e423959daac50674dc" />
Functions and Sensors of Smart Walkers From 2015 to 2024: Scoping Review
Background: Early mobilization and mobility are essential components of the recovery process following surgery and trauma-related hospitalization. In addition to personalized support from physiotherapists and health care professionals, assistive devices such as walkers play a crucial role in facilitating safe and effective mobility. Objective: This scoping review aims to provide a comprehensive overview of the current state of the literature on the design, sensor technologies, and functional applications of smart walkers and to assess the extent to which existing studies reflect clinical use cases. Methods: Peer-reviewed English articles published between 2015 and 2024 were identified by searching PubMed, CINAHL, SSCI, and IEEE, focusing on the topic of smart walkers. Secondary analyses and walkers with 2 wheels or fewer were excluded in abstract screening. Study screening and selection were performed according to the Joanna Briggs Institute guidelines for scoping research and reported following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines. The Rayyan systematic review management software was used for study selection. The articles included were analyzed with respect to the sensor technologies used, their functional capabilities, and their application scenarios. Results: Of the 800 articles screened, 44 (5.5%) met the inclusion criteria. Most of these articles were research reports (n=36, 81.8%) and were conducted in laboratory-based environments (n=30, 68.2%). Most studies evaluated smart walkers in asymptomatic populations (n=29, 65.9%), with half (n=22, 50%) involving younger adults. Among the sensor modalities reported, camera-based and light detection and ranging–based sensors were most prevalent for half of the implementations. Light detection and ranging–based sensors can be categorized according to their primary functions: gait analysis (n=11, 25%), collision detection (n=9, 36%), and navigation (n=5, 11.4%). Load sensors (n=10, 22.7%) and ultrasonic sensors (n=11, 25%) were among the most frequently cited sensor modalities in the literature. Load sensors, also known as force sensors, are integrated into the handlebars, frame, forearm supports, or chest pads of smart walkers. These sensors measure the user’s load, providing essential data for calculating body weight support or inferring the user’s intention to move. Conclusions: The smart walkers described in the literature were predominantly tested in asymptomatic and younger populations. Bridging the gap between current laboratory-based research and real-world clinical environments, as well as the daily lives of end users, remains a critical objective. Addressing the specific needs of older adults through comprehensive requirements analyses and iterative testing continues to be an ongoing challenge, yet these processes can serve as integral components of research and development projects. Trial Registration: OSF Registries osf.io/ctpf4; https://osf.io/ctpf4
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/2832252ade183a1b4c4196b752891efc" />
Efficacy of a World Health Organization–Guided Self-Help Intervention for Reducing Psychological Distress in Afghan Refugees: Randomized Controlled Trial
Background: Common mental health disorders are highly prevalent among refugees. There is an urgent need to address the mental health burden in this population. Objective: This study tested the efficacy of an individually supported self-help stress-management intervention developed by the World Health Organization—Doing What Matters in Times of Stress (DWM)—in reducing psychological distress and improving functioning among refugees in Indonesia, a major transit country in the Asia-Pacific region. Methods: A single-blind randomized controlled trial with 303 Farsi-speaking refugees was conducted between June 2024 and June 2025. Participants with moderate to high psychological distress (Kessler Psychological Distress Scale [K10] score≥20) were randomly allocated to the facilitator-guided individual DWM condition (n=202) or a repeated assessment control condition (n=101). The primary outcome was psychological distress (K10 score) at the posttreatment assessment. Secondary outcomes were posttraumatic stress disorder symptoms, functional impairment, social functioning, and personally identified problems. Results: Intent-to-treat analysis indicated that participants in the DWM condition showed greater reductions in K10 scores than those in the repeated assessment control condition (posttreatment: =−.563, SE=0.124; <.001; Cohen =0.56; 1-month follow-up: =−.447, SE=0.140; =.002; Cohen =0.45). Similarly, those participants in the DWM condition reported greater improvements in posttraumatic stress disorder symptoms, well-being, social functioning, functional impairment, and personally identified psychological problems. No serious adverse events were reported. Conclusions: The findings provide the first evidence for the effectiveness of DWM in reducing psychological distress and improving overall functioning among urban refugees living in a transit setting. Individually supported self-help interventions such as DWM mayoffer an effective, feasible, and scalable approach to improving mental health for refugees. Trial Registration: ANZCTR ACTRN12624000609550; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=387637&isReview=true
Codon Optimization Isn’t Equal: Benchmarking Gene Design for Antibody Expression
Justin Byers
Founder and CEO
Axio BioPharma
Panelist
Justin Byers
Justin Byers is the founder and CEO of Axio BioPharma. He holds a BS in biochemistry and molecular biology from Illinois State University and has held leadership roles at Illumina, Danaher, and Fujifilm. Throughout his career, Byers has led commercial, operational, and cross-functional initiatives supporting biologics programs from early development through manufacturing. He has worked closely with scientific teams to scale workflows, improve process rigor, and align technical execution with strategic objectives. At Axio, Byers oversees corporate strategy, partnerships, and scientific direction. His focus is positioning the company at the intersection of structured data and biologics workflow execution. Axio is accelerating biologics development through mAb production services for R&D while partnering with innovators and CDMOs to ensure the data required for rigorous decision making and a digitally enabled future is generated, structured, and accessible.
Daniel Lin-Arlow, PhD
Chief Scientific Officer and Co-founder
Ansa Biotechnologies
Panelist
Daniel Lin-Arlow, PhD
Daniel Lin-Arlow, PhD, is a scientist-entrepreneur with deep expertise in synthetic biology and biophysics. Motivated by firsthand challenges in obtaining DNA constructs for metabolic engineering in graduate school, he is deeply committed to providing scientists with the DNA constructs they need for their research. As Ansa’s founding CEO, Lin-Arlow grew the company from two employees in 2018 to more than 70 by 2024, raising over $130 million in venture capital and grant funding to support technology development and commercialization. He transitioned to the role of chief scientific officer in 2024, where he leads the development of new applications of the company’s technologies. Lin-Arlow received his PhD from the University of California, Berkeley for his work in Jay Keasling’s lab for developing the DNA synthesis technology commercialized by Ansa. Prior to graduate school, he was a scientific associate at D.E. Shaw Research where he studied the biophysical properties of G protein-coupled receptors, including how drugs bind and modulate their activity. Dan began his scientific career at MIT, where he earned dual SB degrees in math with computer science and biology, and developed computation tools for the analysis of regulation of gene expression at the Broad Institute of MIT and Harvard. Lin-Arlow is a co-inventor of nine patent families and has co-authored scientific publications in Nature, Science, Cell, PNAS, and Nature Biotechnology.
Broadcast Date:
- Time:
Antibody expression titers are key drivers of screening efficiency in discovery, developability, manufacturing economics, and development timelines. Although it is possible to address poor antibody expression by increasing overall batch size and optimizing downstream processes, the root cause often lies in the underlying DNA sequences. Controlled benchmarking studies are helpful for systematically evaluating DNA construct design decisions that impact titers.
In this GEN webinar, Justin Byers and Daniel Lin-Arlow, PhD, examine how enzymatic DNA synthesis and DNA construct design mitigate antibody expression challenges.
Byers will walk through a controlled benchmarking study of codon-optimization approaches, including details of the study design and how structured, gene-to-protein workflows can help identify optimal constructs before they become downstream problems. He will show that under matched CHO and HEK293 conditions, antibody constructs codon-optimized with an AI codon language model had consistently higher transient expression titers than other approaches. The AI codon-optimized sequences contained “complex” features such as repeats and GC skew that challenge traditional gene synthesis processes but were readily manufactured by Ansa’s DNA synthesis platform. These results suggest that complex sequence features can be important for optimal gene expression, which makes the ability to manufacture them as relevant as the codon strategy.
Lin-Arlow will present Ansa’s enzymatic DNA synthesis technology and the benefits to clients working on antibody production, cell and gene therapies, and other synthetic biology applications. Key takeaways include:
- An AI-powered codon optimization strategy that measurably improves transient antibody expression yield
- Why controlled side-by-side benchmarking under standardized conditions is the only reliable way to objectively evaluate DNA construct design choices
- How integrating rigorous sequence evaluation upstream compresses timelines and reduces the risks of expression failures late in development
- How Ansa’s fully enzymatic DNA synthesis addresses complex sequences, including: High or low GC content, secondary structures, inverted terminal repeats (ITRs), and homopolymers
- The Ansa On-Time Guarantee—DNA orders shipped on time, or the complete order is free
A live Q&A session will follow the presentation offering you a chance to pose questions to our expert panelists.
Produced with support from:
The post Codon Optimization Isn’t Equal: Benchmarking Gene Design for Antibody Expression appeared first on GEN – Genetic Engineering and Biotechnology News.
Potato pests
Nature Biotechnology, Published online: 18 May 2026; doi:10.1038/s41587-026-03133-0
Potato pests
Intellia heads to FDA with first in vivo CRISPR-based gene editing therapy
Nature Biotechnology, Published online: 18 May 2026; doi:10.1038/s41587-026-03154-9
Intellia heads to FDA with first in vivo CRISPR-based gene editing therapy
An in vivo electromagnetic field-responsive gene switch
Nature Biotechnology, Published online: 18 May 2026; doi:10.1038/s41587-026-03147-8
An in vivo electromagnetic field-responsive gene switch
STAT+: Maryland state affordability board places a price cap on Ozempic
The Maryland Prescription Drug Affordability Board agreed to set an upper payment limit for the Ozempic diabetes treatment, marking the second time that the state panel has taken such a step in recent weeks.
The board, which is designed to function like a state utility commission, will now oversee a process to lower the cost of the type 2 diabetes medicine for the state and local governments by January 2027. At the time, the price will be capped at $274 for a 30-day supply, a move the board estimates will save $5.8 million a year.
The expected cost was benchmarked against the maximum fair price paid by Medicare, said Andrew York, executive director of the Maryland board. Meanwhile, the board is expected to begin acting in 2028 to set upper payment limits on high-cost drugs purchased by all Marylanders in the commercial insurance market as well.
STAT+: White House taps Amazon, GoodRx, and Mark Cuban to bolster TrumpRx
WASHINGTON — The White House announced an expansion of its prescription drug discount platform, TrumpRx, on Monday, adding more than 600 generic drugs to the website.
The expansion comes via a partnership with entrepreneur Mark Cuban’s Cost Plus Drug Company, as well as Amazon Pharmacy and GoodRx, which both also provide drug savings or low-cost prescription medicines.
“By incorporating this massive catalog of low-cost generics at TrumpRx.Gov, consumers will now have one source to ensure that they’re getting the lowest possible cost on their prescriptions,” President Trump said Monday at a White House event. “They have something that they’ve never had before.”
With no approved vaccine for Ebola outbreak, experts weigh testing a long shot
The latest Ebola outbreak in the Democratic Republic of the Congo, which was only confirmed to be underway at the end of last week, is already the fourth largest on record. The deadly virus is spreading in a conflict zone where recent Ebola experience has shown containment will be a challenge. There is no vaccine that targets the species of the virus that is spreading there, Bundibugyo.
But there is a tiny bit of scientific evidence that suggests the existing licensed Ebola vaccine, Merck’s Ervebo, might offer some protection against this virus, even though it is designed to target a different species of Ebola, Zaire ebolavirus.