Return of Results in Genomic Research Using Large-Scale or Whole Genome Sequencing: Toward a New Normal.

Genome sequencing is increasingly used in research and integrated into clinical care. In the research domain, large-scale analyses, including whole genome sequencing with variant interpretation and curation, virtually guarantee identification of variants that are pathogenic or likely pathogenic and actionable. Multiple guidelines recommend that findings associated with actionable conditions be offered to research participants in order to demonstrate respect for autonomy, reciprocity, and participant interests in health and privacy. Some recommendations go further and support offering a wider range of findings, including those that are not immediately actionable. In addition, entities covered by the US Health Insurance Portability and Accountability Act (HIPAA) may be required to provide a participant's raw genomic data on request. Despite these widely endorsed guidelines and requirements, the implementation of return of genomic results and data by researchers remains uneven. This article analyzes the ethical and legal foundations for researcher duties to offer adult participants their interpreted results and raw data as the new normal in genomic research.

"Choice of law" in precision medicine research.

Large-scale precision medicine research requires massive amounts of data representing people from all walks of life; thus, in the US, it is often multistate research. Significant legal and ethical quandaries arise as a result of the patchwork of laws states have enacted that may apply to research, are not preempted by federal law, and may impose requirements or provide participant rights and protections that differ from other states. Determining which state's laws apply, and under what circumstances, is not solved by the transition to a single-IRB model and researchers cannot simply choose one state's laws to apply uniformly. At a minimum, the current process of meeting each state's requirements could be made more reliable and efficient. To fundamentally change this status quo, however, requires action at multiple levels. Federally, well-known gaps in the Genetic Information Nondiscrimination Act should be closed, and a coherent system of compensation for research injury-including non-physical injuries-should be developed. States should clarify which of their laws are intended to apply to research and work collaboratively to harmonize them. At the level of individual research projects, numerous policies and procedures could be standardized through authoritative guidelines. Examples include clarifying the scope of broad consent, understanding and upholding Certificates of Confidentiality, offering individual research results responsibly, and consistently disseminating aggregate results to participants and the public. Overall, development of a choice of law framework specific to the research context could significantly promote clarity and consistency.

The Ottawa Statement implementation guidance document for cluster randomized trials in the hemodialysis setting.

Research teams are increasingly interested in using cluster randomized trial (CRT) designs to generate practice-guiding evidence for in-center maintenance hemodialysis. However, CRTs raise complex ethical issues. The Ottawa Statement on the Ethical Design and Conduct of Cluster Randomized Trials, published in 2012, provides 15 recommendations to address ethical issues arising within 7 domains: justifying the CRT design, research ethics committee review, identifying research participants, obtaining informed consent, gatekeepers, assessing benefits and harms, and protecting vulnerable participants. But applying the Ottawa Statement recommendations to CRTs in the hemodialysis setting is complicated by the unique features of the setting and population. Here, with the help of content experts and patient partners, we co-developed this implementation guidance document to provide research teams, research ethics committees, and other stakeholders with detailed guidance on how to apply the Ottawa Statement recommendations to CRTs in the hemodialysis setting, the result of a 4-year research project. Thus, our work demonstrates how the voices of patients, caregivers, and all stakeholders may be included in the development of research ethics guidance.

Decision-making and autonomy among participants in early-phase cancer immunotherapy trials: a qualitative study.

BACKGROUND: Participants considering early-phase cancer clinical trials (CTs) need to understand the unique risks and benefits prior to providing informed consent. This qualitative study explored the factors that influence patients' decisions about participating in early-phase cancer immunotherapy CTs through the ethical lens of relational autonomy. METHODS: Using an interpretive descriptive design, interviews were conducted with 21 adult patients with advanced cancer who had enrolled in an early-phase CT. Data was analyzed using relational autonomy ethical theory and constant comparative analysis. RESULTS: The extent to which participants perceived themselves as having a choice to participate in early-phase cancer immunotherapy CTs was a central construct. Perceptions of choice varied according to whether participants characterized their experience as an act of desperation or as an opportunity to receive a novel treatment. Intersecting psychosocial and structural factors influenced participants' decision making about participating in early-phase cancer immunotherapy trials. These relational factors included: (1) being provided with hope; (2) having trust; (3) having the ability to withdraw; and (4) timing constraints. CONCLUSIONS: Findings highlight the continuum of perceived choice that exists among patients with cancer when considering participation in early-phase cancer immunotherapy CTs. All participants were interpreted as exhibiting some degree of relational autonomy within the psychosocial and structural context of early-phase CT decision making. This study offers insights into the intersection of cancer care delivery, personal beliefs and values, and established CT processes and structures that can inform future practices and policies associated with early-phase cancer immunotherapy CTs to better support patients in making informed decisions.

The ethics of induction of labor at 39 weeks in low-risk nulliparas in research and clinical practice.

The "A Randomized Trial of Induction Versus Expectant Management" trial (ARRIVE trial) published in 2018 suggested that induction of labor can be considered a "reasonable option" for low-risk nulliparous women at ≥39 weeks of gestation. The study results led some professional societies to endorse the option for elective induction of labor at 39 weeks of gestation in low-risk nulliparas, and this has begun to change obstetrical practice. The ARRIVE trial provided valuable information supporting the benefits of induction of labor; however, the trial is insufficient to serve as the primary justification for widespread elective induction of labor at 39 weeks of gestation in low-risk nulliparas because of concerns about external validity. Thus, the French ARRIVE trial was designed to test the hypothesis in a different setting that elective induction of labor at 39 weeks of gestation in low-risk nulliparas leads to a lower cesarean delivery rate than expectant management. This ongoing trial has been criticized as "pseudoscientific" and telling "women where, when, and how to give birth." We reject these allegations and extensively examine the ethical framework that should govern clinical and research interventions, including elective induction of labor at 39 weeks of gestation in low-risk nulliparas. This study aimed to discuss the ethical issues that emerge from randomized trials of elective induction of labor at 39 weeks of gestation in low-risk nulliparas and the ethics of the clinical practice itself. The analysis of existing evidence shows the importance of further research on induction of labor at 39 weeks of gestation in low-risk women. Certain aspects of research ethics in this area, particularly the consent of pregnant women in a context where autonomy remains fragile, call for vigilance. In addition, we emphasize that childbirth is not only a medical object but also a social phenomenon that cannot be regarded only from the perspective of a health risk to be managed by clinical research. Further research on this issue is needed to allow pregnant women to make informed decisions, and the results should be integrated with social issues. The perspective of women is required in constructing, evaluating, and implementing medical interventions in childbirth, such as induction of labor at 39 weeks of gestation.

Human vs machine: identifying ChatGPT-generated abstracts in Gynecology and Urogynecology.

BACKGROUND: ChatGPT, a publicly available artificial intelligence large language model, has allowed for sophisticated artificial intelligence technology on demand. Indeed, use of ChatGPT has already begun to make its way into medical research. However, the medical community has yet to understand the capabilities and ethical considerations of artificial intelligence within this context, and unknowns exist regarding ChatGPT's writing abilities, accuracy, and implications for authorship. OBJECTIVE: We hypothesize that human reviewers and artificial intelligence detection software differ in their ability to correctly identify original published abstracts and artificial intelligence-written abstracts in the subjects of Gynecology and Urogynecology. We also suspect that concrete differences in writing errors, readability, and perceived writing quality exist between original and artificial intelligence-generated text. STUDY DESIGN: Twenty-five articles published in high-impact medical journals and a collection of Gynecology and Urogynecology journals were selected. ChatGPT was prompted to write 25 corresponding artificial intelligence-generated abstracts, providing the abstract title, journal-dictated abstract requirements, and select original results. The original and artificial intelligence-generated abstracts were reviewed by blinded Gynecology and Urogynecology faculty and fellows to identify the writing as original or artificial intelligence-generated. All abstracts were analyzed by publicly available artificial intelligence detection software GPTZero, Originality, and Copyleaks, and were assessed for writing errors and quality by artificial intelligence writing assistant Grammarly. RESULTS: A total of 157 reviews of 25 original and 25 artificial intelligence-generated abstracts were conducted by 26 faculty and 4 fellows; 57% of original abstracts and 42.3% of artificial intelligence-generated abstracts were correctly identified, yielding an average accuracy of 49.7% across all abstracts. All 3 artificial intelligence detectors rated the original abstracts as less likely to be artificial intelligence-written than the ChatGPT-generated abstracts (GPTZero, 5.8% vs 73.3%; P<.001; Originality, 10.9% vs 98.1%; P<.001; Copyleaks, 18.6% vs 58.2%; P<.001). The performance of the 3 artificial intelligence detection software differed when analyzing all abstracts (P=.03), original abstracts (P<.001), and artificial intelligence-generated abstracts (P<.001). Grammarly text analysis identified more writing issues and correctness errors in original than in artificial intelligence abstracts, including lower Grammarly score reflective of poorer writing quality (82.3 vs 88.1; P=.006), more total writing issues (19.2 vs 12.8; P<.001), critical issues (5.4 vs 1.3; P<.001), confusing words (0.8 vs 0.1; P=.006), misspelled words (1.7 vs 0.6; P=.02), incorrect determiner use (1.2 vs 0.2; P=.002), and comma misuse (0.3 vs 0.0; P=.005). CONCLUSION: Human reviewers are unable to detect the subtle differences between human and ChatGPT-generated scientific writing because of artificial intelligence's ability to generate tremendously realistic text. Artificial intelligence detection software improves the identification of artificial intelligence-generated writing, but still lacks complete accuracy and requires programmatic improvements to achieve optimal detection. Given that reviewers and editors may be unable to reliably detect artificial intelligence-generated texts, clear guidelines for reporting artificial intelligence use by authors and implementing artificial intelligence detection software in the review process will need to be established as artificial intelligence chatbots gain more widespread use.

Close contacts of xenograft recipients: Ethical considerations due to risk of xenozoonosis.

With decades of pre-clinical studies culminating in the recent clinical application of xenotransplantation, it would appear timely to provide recommendations for operationalizing oversight of xenotransplantation clinical trials. Ethical issues with clinical xenotransplantation have been described for decades, largely centering on animal welfare, the risks posed to the recipient, and public health risks posed by potential spread of xenozoonosis. Much less attention has been given to considerations relating to potentially elevated risks faced by those who may care for or otherwise have close contact with xenograft recipients. This paper examines the ethical and logistical issues raised by the potential exposure to xenozoonotic disease faced by close contacts of xenotransplant recipients-defined herein as including but not limited to caregivers, household contacts, and sexual partners-which warrants special attention given their increased risk of exposure to infection compared to the general public. We discuss implications of assent or consent by these close contacts to potentially undergo, along with the recipient, procedures for infection screening and possible quarantine. We then propose several options and recommendations for operationalizing oversight of xenotransplantation clinical trials that could account for and address close contacts' education on and agency regarding the risk of xenozoonosis.

Advance Consent in Acute Stroke Trials: Survey of Canadian Stroke Physicians.

Advance consent presents a potential solution to the challenge of obtaining informed consent for participation in acute stroke trials. Clinicians in stroke prevention clinics are uniquely positioned to identify and seek consent from potential stroke trial participants. To assess the acceptability of advance consent to Canadian stroke clinic physicians, we performed an online survey. We obtained 58 respondents (response rate 35%): the vast majority (82%) expressed comfort with obtaining advance consent and 92% felt that doing so would not be a significant disruption to clinic workflow. These results support further study of advance consent for acute stroke trials.

Advance Consent in Acute Stroke Trials: Survey of Canadian Research Ethics Board Chairs.

Advance consent could allow individuals at high risk of stroke to provide consent before they might become eligible for enrollment in acute stroke trials. This survey explores the acceptability of this novel technique to Canadian Research Ethics Board (REB) chairs that review acute stroke trials. Responses from 15 REB chairs showed that majority of respondents expressed comfort approving studies that adopt advance consent. There was no clear preference for advance consent over deferral of consent, although respondents expressed significant concern with broad rather than trial-specific advance consent. These findings shed light on the acceptability of advance consent to Canadian ethics regulators.

Impact of institutional review board stipulations on turnaround time for approvals of surgical studies.

BACKGROUND: We have examined the number and types of stipulations received following the submission of surgical study protocols to the Institutional Review Boards (IRBs) for review, and their effect on turnaround time for approval. This analysis will enable our organization to improve the quality of applications and design of study protocols, which can streamline the approval process and increase efficiency of the startup phase for clinical research. METHODS: IRB stipulations for 48 surgical studies were analyzed. Various factors were assessed: surgical specialty, type of study by design, clinical trial phase, type of investigational product, type of IRB utilized (local or centralized), study complexity score, type of review (e.g., exempt, expedited, or full board), turnaround time, and number of stipulations received. Statistical analyses were performed to examine associations between the number/type of stipulations received during the IRB review process and any of the aforementioned study-related factors. RESULTS: For analyzed surgical studies, the number of stipulations allotted to a study and time taken for approval had moderate association with the complexity of the study. The turnaround time for approval was the highest for randomized, controlled trials and studies undergoing full board review. CONCLUSION: This study elucidates characteristics that are associated with increased time for IRB approval. Analysis of IRB stipulations can help improve the turnaround time for the approval process, increase efficiency of startup phase, and transition to execution phase faster, which will allow more time for enrollment of research subjects, and increase return on investment made into research and development programs.

The impact of feedback training on prediction of cancer clinical trial results.

INTRODUCTION: Funders must make difficult decisions about which squared treatments to prioritize for randomized trials. Earlier research suggests that experts have no ability to predict which treatments will vindicate their promise. We tested whether a brief training module could improve experts' trial predictions. METHODS: We randomized a sample of breast cancer and hematology-oncology experts to the presence or absence of a feedback training module where experts predicted outcomes for five recently completed randomized controlled trials and received feedback on accuracy. Experts then predicted primary outcome attainment for a sample of ongoing randomized controlled trials. Prediction skill was assessed by Brier scores, which measure the average deviation between their predictions and actual outcomes. Secondary outcomes were discrimination (ability to distinguish between positive and non-positive trials) and calibration (higher predictions reflecting higher probability of trials being positive). RESULTS: A total of 148 experts (46 for breast cancer, 54 for leukemia, and 48 for lymphoma) were randomized between May and December 2017 and included in the analysis (1217 forecasts for 25 trials). Feedback did not improve prediction skill (mean Brier score for control: 0.22, 95% confidence interval = 0.20-0.24 vs feedback arm: 0.21, 95% confidence interval = 0.20-0.23; p = 0.51). Control and feedback arms showed similar discrimination (area under the curve = 0.70 vs 0.73, p = 0.24) and calibration (calibration index = 0.01 vs 0.01, p = 0.81). However, experts in both arms offered predictions that were significantly more accurate than uninformative forecasts of 50% (Brier score = 0.25). DISCUSSION: A short training module did not improve predictions for cancer trial results. However, expert communities showed unexpected ability to anticipate positive trials.Pre-registration record: https://aspredicted.org/4ka6r.pdf.

Ethical approval for controlled human infectious model clinical trial protocols - A workshop report.

Controlled Human Infectious Model studies (CHIM) involve deliberately exposing volunteers to pathogens. To discuss ethical issues related to CHIM, the European Vaccine Initiative and the International Alliance for Biological Standardization organised the workshop "Ethical Approval for CHIM Clinical Trial Protocols", which took place on May 30-31, 2023, in Brussels, Belgium. The event allowed CHIM researchers, regulators, ethics committee (EC) members, and ethicists to examine the ethical criteria for CHIM and the role(s) of CHIM in pharmaceutical development. The discussions led to several recommendations, including continued assurance that routine ethical requirements are met, assurance that participants are well-informed, and that preparation of study documents must be both ethically and scientifically sound from an early stage. Study applications must clearly state the rationale for the challenge compared to alternative study designs. ECs need to have clear guidance and procedures for evaluating social value and assessing third-party risks. Among other things, public trust in research requires minimisation of harm to healthy volunteers and third-party risk. Other important considerations include appropriate stakeholder engagement, public education, and access to health care for participants after the study.

Hard to reach? Methodological challenges researching vulnerable, gang-involved, young people.

INTRODUCTION: Research with young people (YP) is ethically challenging and bound in a complex maze of issues relating to power, voice and representation. Such sensitivities mean that the challenges raised in researching marginalised YP are often hard to navigate. This paper reports on research carried out with YP to explore links between mental health, school exclusion and involvement in criminal gangs. It aims to provide a practical guide to negotiating some of the methodological and ethical challenges experienced. METHOD: In-depth interviews conducted with 28 YP (aged 14-24 years) who were gang involved or seen to be at risk of gang involvement. Research was conducted in youth clubs, alternative provision and youth justice settings. RESULTS: OBSERVATIONS/REFLECTIONS: We reflect on how navigating ethics can create barriers to involving YP as primary informants in research. We consider why it is important to overcome these hurdles and how public engagement work with recognised gatekeepers and the use of creative interview methods can facilitate meaningful encounters, where YP feel able to share valuable insights into their lives. CONCLUSION: Alongside a number of specific learning points, the paper reflects on theories behind research with YP, including the need for recognition of power imbalances and reflexivity. It concludes with thoughts on the practical realities of achieving meaningful participation or an 'authentic voice' with marginalised groups and the importance of this in informing policy and practice. PATIENT OR PUBLIC CONTRIBUTION: The focus of this work was to collect experiences of YP who are recognised as gang-involved or at risk of being so, with a view to informing health and education policies. The scoping study for the project involved extensive public engagement work with YP exploring and trialling suitable methods of accessing, recruiting and ultimately interviewing this target group. This is central to the discussion within the body of the paper.

Returning Individual Research Results from Digital Phenotyping in Psychiatry.

Psychiatry is rapidly adopting digital phenotyping and artificial intelligence/machine learning tools to study mental illness based on tracking participants' locations, online activity, phone and text message usage, heart rate, sleep, physical activity, and more. Existing ethical frameworks for return of individual research results (IRRs) are inadequate to guide researchers for when, if, and how to return this unprecedented number of potentially sensitive results about each participant's real-world behavior. To address this gap, we convened an interdisciplinary expert working group, supported by a National Institute of Mental Health grant. Building on established guidelines and the emerging norm of returning results in participant-centered research, we present a novel framework specific to the ethical, legal, and social implications of returning IRRs in digital phenotyping research. Our framework offers researchers, clinicians, and Institutional Review Boards (IRBs) urgently needed guidance, and the principles developed here in the context of psychiatry will be readily adaptable to other therapeutic areas.

First-in-Human Whole-Eye Transplantation: Ensuring an Ethical Approach to Surgical Innovation.

As innovations in the field of vascular composite allotransplantation (VCA) progress, whole-eye transplantation (WET) is poised to transition from non-human mammalian models to living human recipients. Present treatment options for vision loss are generally considered suboptimal, and attendant concerns ranging from aesthetics and prosthesis maintenance to social stigma may be mitigated by WET. Potential benefits to WET recipients may also include partial vision restoration, psychosocial benefits related to identity and social integration, improvements in physical comfort and function, and reduced surgical risk associated with a biologic eye compared to a prosthesis. Perioperative and postoperative risks of WET are expected to be comparable to those of facial transplantation (FT), and may be similarly mitigated by immunosuppressive protocols, adequate psychosocial support, and a thorough selection process for both the recipient and donor. To minimize the risks associated with immunosuppressive medications, the first attempts in human recipients will likely be performed in conjunction with a FT. If first-in-human attempts at combined FT-WET prove successful and the biologic eye survives, this opens the door for further advancement in the field of vision restoration by means of a viable surgical option. This analysis integrates recent innovations in WET research with the existing discourse on the ethics of surgical innovation and offers preliminary guidance to VCA programs considering undertaking WET in human recipients.

Which Benefits Can Justify Risks in Research?

Research ethics committees (RECs) evaluate whether the risk-benefit ratio of a study is acceptable. Decentralized clinical trials (DCTs) are a novel approach for conducting clinical trials that potentially bring important benefits for research, including several collateral benefits. The position of collateral benefits in risk-benefit assessments is currently unclear. DCTs raise therefore questions about how these benefits should be assessed. This paper aims to reconsider the different types of research benefits, and their position in risk-benefit assessments. We first propose a categorization of research benefits, based on the types of benefits that can be distinguished from the literature and ethical guidelines. Secondly, we will reconsider the position of collateral benefits. We argue that these benefits are not fundamentally different from other benefits of research and can therefore be included in risk-benefit assessments of DCTs.

Virtue ethics and the unsettled ethical questions in controlled human infection studies.

Controlled human infection studies (CHIs) involve the intentional infection of human subjects for a scientific aim. Though some past challenge trials have involved serious ethical abuses, in the last few decades, CHIs have had a strong track record of safety. Despite increased attention to the ethics of CHIs during the COVID-19 pandemic, CHIs remain controversial, and there has been no in-depth treatment of CHIs through the lens of virtue ethics. In this article, we argue that virtue theory can be helpful for addressing CHIs that present a constellation of controversial, unresolved, and/or under-regulated ethical issues. We begin with some brief background on virtue ethics. We then substantiate our claim that some CHIs raise a constellation of ethical issues that are unresolved in the ethics literature and/or lack adequate regulatory guidance by demonstrating that CHIs can present indeterminate social value, risks to third parties, limitations on the right to withdraw from research, and questions about the upper limit of allowable risk. We argue that the presence of a virtuous investigator, with virtues such as prudence, compassion, and integrity, is especially important when these unresolved research ethics issues arise, which is the case for certain types of controlled human infection studies. We use the historical example of Walter Reed and the Yellow Fever Commission to illustrate this claim, and we also highlight some contemporary examples. We end by sketching some practical implications of our view, such as ensuring that investigators with experience running CHIs are involved in novel CHI models.

Why restrict medical effective altruism?

In a challenge trial, research subjects are purposefully exposed to some pathogen in a controlled setting, in order to test the efficacy of a vaccine or other experimental treatment. This is an example of medical effective altruism (MEA), where individuals volunteer to risk harms for the public good. Many bioethicists rejected challenge trials in the context of Covid-19 vaccine research on ethical grounds. After considering various grounds of this objection, I conclude that the crucial question is how much harm research subjects can permissibly risk. But we lack a satisfying way of making this judgment that does not appeal simply to the intuitions of doctors or bioethicists. I consider one recent and structurally plausible approach to critically evaluating the harm question. Alex London defends a social consistency test for research risks: we should compare the risks undertaken by research subjects to relevantly similar risks which are accepted in other spheres of society. I argue there is no good reason not to consider volunteer military service as a relevant social comparison. This implies there is essentially no cap on acceptable risks on the social consistency rationale. In short, if soldiers can be heroes, why can't research volunteers?

Therapeutic misunderstandings in modern research.

Clinical trials play a crucial role in generating evidence about healthcare interventions and improving outcomes for current and future patients. For individual trial participants, however, there are inevitably trade-offs involved in clinical trial participation, given that trials have traditionally been designed to benefit future patient populations rather than to offer personalised care. Failure to understand the distinction between research and clinical care and the likelihood of benefit from participation in clinical trials has been termed the 'therapeutic misconception'. The evolution of the clinical trials landscape, including greater integration of clinical trials into healthcare and development of novel trial methodologies, may reinforce the significance of the therapeutic misconception and other forms of misunderstanding while at the same time (paradoxically) challenging its salience. Using cancer clinical trials as an exemplar, we describe how methodological changes in early- and late-phase clinical trial designs, as well as changes in the design and delivery of healthcare, impact upon the therapeutic misconception. We suggest that this provides an impetus to re-examine the ethics of clinical research, particularly in relation to trial access, participant selection, communication and consent, and role delineation.

Research ethics and artificial intelligence for global health: perspectives from the global forum on bioethics in research.

BACKGROUND: The ethical governance of Artificial Intelligence (AI) in health care and public health continues to be an urgent issue for attention in policy, research, and practice. In this paper we report on central themes related to challenges and strategies for promoting ethics in research involving AI in global health, arising from the Global Forum on Bioethics in Research (GFBR), held in Cape Town, South Africa in November 2022. METHODS: The GFBR is an annual meeting organized by the World Health Organization and supported by the Wellcome Trust, the US National Institutes of Health, the UK Medical Research Council (MRC) and the South African MRC. The forum aims to bring together ethicists, researchers, policymakers, research ethics committee members and other actors to engage with challenges and opportunities specifically related to research ethics. In 2022 the focus of the GFBR was "Ethics of AI in Global Health Research". The forum consisted of 6 case study presentations, 16 governance presentations, and a series of small group and large group discussions. A total of 87 participants attended the forum from 31 countries around the world, representing disciplines of bioethics, AI, health policy, health professional practice, research funding, and bioinformatics. In this paper, we highlight central insights arising from GFBR 2022. RESULTS: We describe the significance of four thematic insights arising from the forum: (1) Appropriateness of building AI, (2) Transferability of AI systems, (3) Accountability for AI decision-making and outcomes, and (4) Individual consent. We then describe eight recommendations for governance leaders to enhance the ethical governance of AI in global health research, addressing issues such as AI impact assessments, environmental values, and fair partnerships. CONCLUSIONS: The 2022 Global Forum on Bioethics in Research illustrated several innovations in ethical governance of AI for global health research, as well as several areas in need of urgent attention internationally. This summary is intended to inform international and domestic efforts to strengthen research ethics and support the evolution of governance leadership to meet the demands of AI in global health research.