Monthly News and Updates (July 2026)

In July 2026, AI-in-healthcare regulatory activity centered on clarifying how existing frameworks apply to fast-moving AI tools rather than creating new ones outright. Health Canada issued guidance on AI-generated content in device submissions, HHS committed department-wide resources to AI-driven biomedical research under the White House’s Genesis Mission, and the UK’s MHRA and NHS England jointly clarified when AI “scribe” tools fall inside or outside medical device regulation. Meanwhile, the EU’s AI Omnibus entered into force just days before the AI Act’s original high-risk compliance deadline, extending the timeline for AI embedded in regulated products, including medical devices, by a full year.

Disclaimer: This article is produced for educational and informational purposes only. It summarises publicly available regulatory updates and reflects our interpretation of emerging developments; it does not constitute legal, regulatory, financial, or professional advice, nor should it be interpreted as a statement of regulatory intent. Readers should consult qualified counsel before taking any action based on the information presented here.

Canada

1) Health Canada issues Notice to Industry on AI tools in Class III/IV device submissions (Published: July 16, 2026)

Health Canada published a Notice to Industry addressing a significant increase in the volume of information being submitted with Class III and IV medical device licence applications. The notice reminds applicants that, as set out in the Medical Devices Regulations, applications for Class III and IV licences must include summaries of the studies and evidence submitted, and that these summaries should be clear, concise, and accurately reflect the supporting data. Including unnecessary or duplicative information, the notice states, can increase review times and delay licensing.

Notably, the guidance includes a dedicated section on the use of artificial intelligence tools: if applicants use AI tools to generate summaries or explanatory material for their submissions, they remain responsible for ensuring that content is accurate and reflects the supporting data.

Expectations for information submitted with Class III and IV medical device licence applications: Notice to industry

How it applies to AI in Healthcare:

This is a direct signal that Health Canada is watching for AI-generated content in regulatory submissions themselves, not just in the devices being licensed. Regulatory affairs teams using generative AI tools to help draft submission summaries or explanatory material should treat this as a reminder that responsibility for accuracy cannot be delegated to the tool; any AI-assisted content should go through the same rigorous human verification against source data as any other part of a licence application, and organizations may want to document that verification step as part of their internal quality process.

US

1) HHS joins the White House’s “Genesis Mission” for AI-driven biomedical research (Published: July 22, 2026)

The U.S. Department of Health and Human Services announced it is aligning expertise and capabilities across the Department to support the Genesis Mission, a whole-of-government AI initiative launched by presidential action in November 2025. Working with the White House Office of Science and Technology Policy, HHS will launch a series of National Science and Technology Challenges inviting the scientific community to apply AI to major health problems, with a flagship challenge focused on identifying the root causes of chronic disease, and additional challenges targeting pediatric cancer research and drug discovery and development.

Key Points

• NIH-led Bio Genesis Mission. The centerpiece of HHS’s participation is the NIH-led Bio Genesis Mission, which aims to use AI, advanced computing, and cross-sector partnerships to modernize the biomedical research ecosystem, with a stated goal of cutting in half the time it takes for scientific discoveries to reach patients.

• Broad dataset contribution. HHS agencies including NIH, FDA, and ARPA-H will contribute scientific resources such as biomedical research, genomics, public health, and safety data, which will be combined with the Genesis Mission’s advanced computing capabilities.

• Part of a larger federal commitment. The announcement was made alongside a broader White House rollout of federal commitments to the Genesis Mission, with more than a dozen agencies contributing research awards, funding opportunities, datasets, and research facilities across the initiative.

HHS Joins the Genesis Mission to End America’s Chronic Disease Epidemic

How it applies to AI in Healthcare:

This signals substantial federal appetite for AI-driven biomedical research partnerships, and it may open new funding, data-access, or collaboration opportunities for organizations working in chronic disease, pediatric cancer, or drug discovery. Companies and research institutions in these spaces may wish to monitor the National Science and Technology Challenges as they are formally launched, since participation could offer access to federal datasets and computing infrastructure that would otherwise be difficult to obtain.

EU & UK

1) MHRA and NHS England clarify medical device status of AI “scribes” (Published: July 29, 2026)

The Medicines and Healthcare products Regulatory Agency (MHRA), working closely with NHS England, published guidance clarifying how existing medical device law applies to ambient voice technology (AVT) products, commonly known as AI scribes, used in health and care settings in Great Britain. The guidance does not change the underlying law; it explains how the MHRA applies existing medical device regulations to this technology.

Key Points

• What is not regulated as a medical device. AVT products intended solely for transcription, summarising clinical conversations, drafting letters, or suggesting clinical codes for a clinician to review are not regulated as medical devices under the current framework.

• What is regulated as a medical device. AVT products intended to support diagnosis, treatment, or prevention, or that take automated action such as placing orders without clinician review, are regulated as medical devices and must meet the relevant safety and performance requirements.

• Clinician responsibility is unchanged. Clinicians remain responsible for reviewing and verifying AI-generated transcripts, summaries, and other outputs before they are used in patient care, regardless of a product’s regulatory classification.

• Governance expectations for NHS organizations. NHS boards and executive teams are asked to ensure AVT deployment is supported by appropriate clinical oversight, local governance, staff training, and procurement processes, and that any change in a product’s functionality that may alter its regulatory status is identified and reviewed.

MHRA clarifies regulatory status of ambient voice technologies used in the NHS

How it applies to AI in Healthcare:

This resolves a significant source of ambiguity that had reportedly slowed AVT adoption across the NHS, and it gives both suppliers and healthcare organizations a clearer basis for classifying their products and deployment models. Vendors building ambient scribing tools should review their intended purpose statements, labelling, and marketing claims against the guidance’s examples, since language suggesting diagnostic or treatment support (rather than pure transcription and drafting) could shift a product into medical device territory. Healthcare organizations adopting these tools should ensure clinical oversight, governance, and staff training processes are documented and kept current as product functionality evolves.

2) NHS England publishes updated guidance on AI-enabled ambient scribing products (Updated: July 29, 2026)

Alongside the MHRA clarification, NHS England published an updated (version 3) long-read guidance document for those adopting AI-enabled ambient scribing products across health and care settings in England. The guidance is aimed at Chief Information Officers, Chief Clinical Information Officers, and technical and product teams leading AI adoption, and it was updated to align with the new MHRA position on regulatory status.

Key Points

• Structured implementation guidance. The document sets out considerations across risk identification, regulatory compliance, data protection and cybersecurity, integration and performance, portability and scalability, support and maintenance, procurement, monitoring and bias mitigation, user training, and liability.

• AVT Supplier Registry. The guidance complements NHS England’s national self-certified AVT Supplier Registry, which lists suppliers that have evidenced criteria including integration capability, performance monitoring, and compliance with the Digital Technology Assessment Criteria (DTAC) and MHRA registration where applicable.

• Bias and accuracy considerations. The guidance flags that AVT products may have varying accuracy across accents, regional dialects, non-native English speakers, and speech disorders or impairments, and asks adopting organizations to confirm products work well across their patient populations and to monitor for any groups that hesitate to seek care due to the technology’s use.

• Liability remains with the NHS organization by default. The guidance notes that liability for AI-enabled products in NHS settings is complex and largely uncharted, and that a non-delegable duty of care may leave liability with the Trust or primary care provider absent clear contractual arrangements with suppliers.

Guidance on the use of AI-enabled ambient scribing products in health and care settings

How it applies to AI in Healthcare:

This guidance functions as a practical operating manual for any organization deploying AI scribes in NHS settings, well beyond the regulatory classification question addressed by MHRA. Compliance and clinical safety teams should pay particular attention to the DCB0160 documentation and Data Protection Impact Assessment requirements, the bias and accuracy monitoring expectations, and the liability provisions, since these apply regardless of whether a given product is classified as a medical device.

3) EU “AI Omnibus” enters into force, delaying AI Act deadlines for medical devices (Published: July 27, 2026)

Regulation (EU) 2026/1744, the Digital Omnibus on AI, entered into force on July 27, 2026, three days after publication in the Official Journal, and just days ahead of the AI Act’s original August 2, 2026 general application date. First proposed by the European Commission in November 2025 as part of a broader simplification package, the regulation amends the AI Act’s implementation timeline in response to practical challenges, including delays in designating national competent authorities and conformity assessment bodies, and the absence of finalized harmonized standards for high-risk AI systems.

Key Points

• Medical devices get an extra year. The compliance deadline for high-risk AI systems embedded in products already subject to EU product safety legislation, including AI-enabled medical devices, moves from August 2, 2027 to August 2, 2028.

• Stand-alone high-risk systems also delayed. The deadline for stand-alone high-risk AI systems under Annex III (covering areas such as employment, education, and access to essential services) moves from August 2, 2026 to December 2, 2027.

• Transparency and GPAI obligations unaffected. Article 50 transparency and AI content-marking duties still apply from August 2, 2026, as do general-purpose AI model provider obligations (in effect since August 2025) and the Article 5 prohibited-practices rules (in effect since February 2025).

• Deadlines are now fixed. Unlike the Commission’s original November 2025 draft, which tied the extended deadlines to confirmation that supporting measures such as harmonized standards were in place, the adopted text sets fixed calendar dates that do not depend on further conditions.

AI Omnibus enters into force

How it applies to AI in Healthcare:

This gives manufacturers of AI-enabled medical devices meaningfully more runway, an additional year, to build out AI Act compliance programs, though it does not reduce the scope of what will eventually be required, and it does not touch existing Medical Device Regulation (MDR) or In Vitro Diagnostic Regulation (IVDR) obligations, which remain fully in force on their own timelines. Organizations should treat the extended deadline as planning time rather than a reason to deprioritize AI Act readiness, particularly given that transparency obligations under Article 50 remain unaffected and already apply. Companies with generative or content-producing AI features in health-adjacent products should confirm whether Article 50 marking duties apply to them now, independent of the high-risk timeline extension.

Cross-Cutting Themes

Disclaimer: The cross-cutting themes below are analytical observations based on the sources cited in this article and reflect our interpretation of emerging regulatory and policy developments. They do not constitute legal, regulatory, or professional advice, nor should they be interpreted as statements of regulatory intent. Readers may wish to consult qualified advisors regarding the application of any regulatory requirements to their specific circumstances.

• Regulators are clarifying existing frameworks rather than writing new rules. Health Canada’s AI submission notice, the MHRA/NHS England AVT guidance, and the EU AI Omnibus all work by interpreting or adjusting how existing regulatory structures apply to AI, rather than introducing standalone AI-specific regulatory regimes.
• Accountability for AI outputs consistently sits with the human user. Whether it’s a regulatory affairs professional verifying AI-drafted submission summaries in Canada, or a clinician reviewing AI-generated transcripts in the NHS, this month’s guidance repeatedly places the burden of accuracy and safety on the human in the loop, not the AI tool itself.
• Governments are treating AI as core research infrastructure, not just a compliance topic. HHS’s alignment with the Genesis Mission reflects a broader pattern of governments positioning AI as central to accelerating biomedical discovery, alongside the more compliance-oriented developments seen elsewhere this month.
• Regulatory timelines are proving flexible when implementation infrastructure lags. The EU’s decision to push back AI Act high-risk deadlines, citing the absence of finalized harmonized standards and conformity assessment bodies, illustrates that regulators may adjust ambitious timelines when the supporting ecosystem isn’t ready, though the underlying obligations are deferred rather than removed.

Key Considerations for Regulatory Alignment

For Founders & Business Owners

• Document your AI-assisted regulatory workflows. If your team uses AI tools anywhere in the process of preparing Health Canada submissions, build a documented human-verification step into your process now, ahead of any future audit or review.
• Review AVT product claims against the new MHRA line. If you offer or are building an ambient scribing tool for UK healthcare settings, revisit your intended purpose statement and marketing language to understand which side of the medical device line your product sits on, and what that means for your go-to-market plan.
• Use the extended EU AI Act runway deliberately. The additional year for embedded high-risk systems is a planning opportunity, not a reason to pause; consider using this time to build out your classification inventory and evidence base while demand on external assessors and consultants is comparatively lower.

For Compliance & Regulatory Specialists

• Track the Genesis Mission’s science and technology challenges as they launch. These may represent meaningful funding or data-access opportunities for organizations in chronic disease, pediatric cancer, or drug discovery research, and eligibility criteria will be worth reviewing closely as they are published.
• Separate AI Act transparency duties from high-risk timeline planning. Confirm which of your organization’s AI systems trigger Article 50 marking and disclosure obligations, since these apply now regardless of the high-risk deadline extension, and are easy to overlook when attention shifts to the delayed deadlines.
• Map liability and governance obligations for AVT tools separately from medical device status. NHS England’s guidance on ambient scribing products imposes DCB0160, DPIA, bias-monitoring, and liability-related expectations that apply whether or not a given tool is classified as a medical device; compliance programs should track these as a distinct workstream.

Sources

• Expectations for information submitted with Class III and IV medical device licence applications: Notice to industry (Released: July 16, 2026)
  https://www.canada.ca/en/health-canada/services/drugs-health-products/medical-devices/application-information/guidance-documents/expectations-information-submitted-class-iii-iv-licence-notice-industry.html

• HHS Joins the Genesis Mission to End America’s Chronic Disease Epidemic (Released: July 22, 2026)
  https://www.hhs.gov/press-room/hhs-joins-genesis-mission-ai-chronic-disease-research.html

• MHRA clarifies regulatory status of ambient voice technologies used in the NHS (Released: July 29, 2026)
  https://www.gov.uk/government/news/mhra-clarifies-regulatory-status-of-ambient-voice-technologies-used-in-the-nhs

• Guidance on the use of AI-enabled ambient scribing products in health and care settings (Updated: July 29, 2026)
  https://www.england.nhs.uk/long-read/guidance-on-the-use-of-ai-enabled-ambient-scribing-products-in-health-and-care-settings/

• AI Omnibus enters into force (Released: July 27, 2026)
  https://digital-strategy.ec.europa.eu/en/news/ai-omnibus-enters-force