Digital Health Regulation in 2026: AI Washing, FDA Guidance, and the Changing Compliance Landscape
A review of key digital health regulatory developments in 2026, including AI-washing risks in health tech M&A, FDA's new approach methodologies draft guidance, and the changing enforcement landscape for at-home diagnostics.
Executive Summary
The convergence of artificial intelligence (AI), digital health, and clinical research has created a complex regulatory environment that continues to evolve rapidly. According to the Spring/Summer 2026 edition of Jones Day's Vital Signs: Digital Health Law Update, three areas are drawing particular attention: the rise of AI washing in digital health transactions, the U.S. Food and Drug Administration's (FDA) draft guidance on New Approach Methodologies (NAMs) for drug development, and enforcement questions surrounding at-home diagnostic tests after the vacatur of the laboratory developed test (LDT) rule. These developments underscore the growing importance of data provenance, model transparency, and regulatory alignment in the biotechnology and medical technology sectors.
Introduction
Digital health technologies are no longer peripheral to health care; they are integral to drug development, patient monitoring, diagnostics, and clinical decision-making. As the FDA and international regulators update their frameworks, companies must adapt their compliance strategies, transaction diligence, and research operations. The regulatory signals from the first half of 2026 suggest a continued push toward integrating real-world evidence, AI-enabled tools, and human-centric testing methods into mainstream biomedical innovation.
Scientific Background and Regulatory Context
The modern digital health landscape includes software as a medical device, AI/ML algorithms, decentralized clinical trials, and home-based testing. Regulatory agencies have been working to keep pace with these advances. In the United States, the FDA has issued multiple guidance documents and enforcement decisions that shape how digital health products are developed and brought to market.
A particular area of focus is the status of laboratory developed tests. In March 2025, a federal court vacated the FDA's final rule that would have regulated LDTs as medical devices. The agency subsequently moved to revert to its prior enforcement discretion approach. This legal backdrop has created uncertainty for diagnostics companies, especially those offering direct-to-consumer at-home testing kits.
Research Findings and Regulatory Developments
AI Washing in Digital Health Transactions
One of the most notable industry developments highlighted in the legal update is the problem of AI washing in mergers and acquisitions (M&A). The term describes digital health companies that exaggerate their AI capabilities—presenting basic analytics or API wrappers as proprietary medical AI—in order to justify higher valuations. For buyers, this creates substantial due diligence challenges.
Traditional software audits are insufficient for assessing machine learning systems, whose value lies in numerical parameters derived from training data of varying provenance and legality. The legal update stresses that access to health data through clinical partnerships does not automatically grant the right to train commercial AI models. Buyers must trace the chain of title for training datasets and determine whether the target owns a proprietary foundation model or simply relies on third-party APIs.
Rather than attempting to eliminate all risks, the report recommends a shift toward risk quantification. This involves estimating the magnitude and likelihood of potential legal liabilities and using tailored representations, warranties, and indemnities to allocate those risks. In many cases, remediation is possible: models trained on problematic data can be retrained, biased algorithms can be fine-tuned, and open-source components can be replaced. Ring-fencing and post-closing retraining covenants are practical options for preserving deal value.
FDA Draft Guidance on New Approach Methodologies
On March 18, 2026, the FDA issued draft guidance titled General Considerations for the Use of New Approach Methodologies in Drug Development. This document provides a validation framework for NAMs—a set of innovative testing approaches that include organoids, organs-on-chips, and in silico computational models. These methods are intended to reduce or replace traditional animal studies while improving predictions of human safety and efficacy.
The guidance is grounded in the FDA Modernization Act 2.0, which clarified that NAMs can support new drug applications in lieu of animal testing. The draft guidance proposes four key validation principles: Context of Use, Human Biological Relevance, Technical Characterization, and Fit-for-Purpose. For drug developers, this represents a meaningful step toward regulatory acceptance of more human-relevant testing methods, potentially accelerating timelines and reducing reliance on animal models.
FDA Warning Letter on At-Home Testing Kits
On March 17, 2026, the FDA issued a warning letter to an at-home testing kit manufacturer for offering an HIV serological diagnostic dried blood spot card self-collection kit without FDA marketing authorization. The agency determined that the kit was adulterated and misbranded under the Federal Food, Drug, and Cosmetic Act. The manufacturer had argued that it only facilitates access to professional medical services, but the FDA rejected that characterization.
This enforcement action is significant because it comes after the vacatur of the LDT rule and the agency's decision to return to its previous oversight approach. The warning letter signals that the FDA continues to scrutinize direct-to-consumer diagnostics and that companies should not assume the LDT vacatur creates a regulatory gap for at-home collection devices.
Industry Impact
The regulatory developments of 2026 have broad implications across the biotechnology and health technology industries.
In the M&A arena, AI washing is reshaping transaction strategy. Buyers are now more likely to conduct deep technical due diligence, including audits of training data, model architecture, and third-party dependencies. Sellers, in turn, need to prepare cleaner data provenance documentation and be ready to explain their actual AI capabilities. This may lead to more realistic valuations and a reduction in inflated AI premium deals.
For pharmaceutical and biotech companies, the FDA's NAM guidance offers a potential path to more efficient nonclinical development. If finalized, it could encourage wider use of organoids, computational models, and other human-centric testing platforms. This aligns with broader trends in precision medicine and the growing emphasis on reducing animal testing. However, the draft nature of the guidance means that adoption will depend on case-by-case regulatory acceptance.
Diagnostics companies face an uncertain enforcement landscape. The LDT rule's vacatur provided some relief, but the warning letter demonstrates that the FDA remains active. Companies developing at-home tests need to carefully assess whether their products fall under FDA jurisdiction and whether marketing authorization is required. Additionally, the legal status of LDTs under the new framework continues to evolve, and legislative or regulatory clarification may be forthcoming.
Clinical and Regulatory Perspective
From a clinical perspective, the shift toward NAMs could improve the predictive power of nonclinical studies by using human-relevant models that more closely mirror physiological responses. Yet there are challenges. The validation principles outlined by the FDA require context-specific evidence, and researchers will need to generate robust data demonstrating that NAMs are fit for their intended regulatory purpose. This is still an emerging area, and clinical evidence remains necessary to confirm real-world benefits.
Regarding at-home diagnostics, patient access and convenience are important benefits, but they must be balanced against safety and test accuracy. The FDA's enforcement action underscores the need for companies to work with regulators early in development, especially when products are distributed directly to consumers. The uncertainty about LDT oversight also calls for careful regulatory planning and possibly engagement with the FDA on product-specific questions.
From industry and investment perspectives, these developments emphasize the need for strong compliance infrastructure, transparent AI governance, and rigorous data management. Companies that treat regulatory alignment as a strategic priority are likely to be better positioned for successful M&A exits, clinical approvals, and market adoption.
Future Outlook
Looking ahead to the next decade, several trends are likely to shape digital health and biotechnology regulation.
AI governance will continue to mature. The issue of AI washing will likely diminish as appraisal and diligence practices improve and regulatory agencies develop clearer definitions for AI/ML claims. International frameworks, such as the EU's AI Act and the UK's evolving approach, will add further complexity for global companies. In the United States, we may see more targeted FDA guidance on AI-enabled medical devices and clinical trial software.
The adoption of NAMs is expected to grow, driven by the FDA Modernization Act 2.0 and ongoing research into organ-on-chip and in silico technologies. Over the next 10 to 15 years, nonclinical testing may become increasingly human-centric, with computational simulations playing a larger role in predicting clinical outcomes. This could reduce drug development costs and timelines, though it will require significant investment in validation and regulatory science.
The diagnostics landscape will also evolve. The question of how the FDA should oversee LDTs remains unresolved, and there are likely to be ongoing debates about balancing innovation, patient access, and safety. At-home testing, especially for infectious diseases and chronic conditions, is likely to expand, prompting regulators to establish clearer frameworks for self-collection devices and digital reporting.
Finally, digital health transactions will require more sophisticated due diligence and post-closing integration plans. As AI becomes increasingly embedded in health care products, buyers will seek deals that include not only technology but also clean data assets, robust AI governance, and regulatory clearance pathways. Sellers who proactively address these issues will have a competitive advantage.
Conclusion
The first half of 2026 has demonstrated that digital health regulation is not static. FDA's draft guidance on NAMs opens new possibilities for drug development, while its warning letter on at-home testing shows continued enforcement focus on market access. At the same time, AI washing is becoming a key consideration in health tech M&A, requiring buyers and sellers to approach deals with greater technical and legal rigor. For the biotechnology and life sciences community, these developments reinforce the importance of integrating scientific innovation with regulatory readiness and commercial strategy.
Key Takeaways
- AI washing is a growing due diligence issue in digital health M&A; buyers must scrutinize whether companies have genuine AI capabilities or simply rely on third-party APIs.
- FDA's draft NAM guidance outlines four validation principles and builds on the FDA Modernization Act 2.0, signaling a shift toward human-centric testing methods.
- A March 2026 warning letter to an at-home HIV testing manufacturer shows that the FDA continues to enforce marketing authorization requirements for diagnostics, regardless of LDT rule changes.
- Data provenance and regulatory compliance are critical for any digital health company, especially when training AI models on clinical data.
- Future regulatory clarity will come from ongoing FDA guidance, international AI rules, and new frameworks for diagnostics and clinical research.