EU AI Act Meets FDA: De-Risking Global Health AI Investments

The global landscape for AI in healthcare is rapidly converging, forcing health plan executives and health IT professionals to confront a complex web of dual compliance requirements. The stakes are higher than ever, with regulatory bodies in both the US and Europe scrutinizing AI health tools with increasing intensity. Companies that have proactively architected their solutions with regulatory pathways in mind, particularly the FDA’s Software as a Medical Device (SaMD) framework, are demonstrably better positioned to navigate this new era of oversight, while those without a defined strategy face escalating enforcement and exclusion risks.

Navigating the Dual Regulatory Gauntlet: A Comparative Look at AI Health Innovators

The journey of AI health companies from innovative concept to widespread adoption is increasingly punctuated by rigorous regulatory checkpoints. Understanding how leading players are addressing these challenges provides crucial insight for health plans and IT departments evaluating integration.

Tempus AI: Precision Medicine at Scale

Tempus AI operates at the intersection of AI and precision medicine, primarily focusing on oncology. Their model involves collecting and analyzing large datasets of clinical and molecular data to assist physicians in making data-driven treatment decisions. This includes genomic sequencing, real-world evidence, and AI-powered analytics to personalize cancer care. Tempus’s approach targets improved diagnostic accuracy, treatment selection, and ultimately, patient outcomes in complex diseases. Their commercial strategy emphasizes partnerships with healthcare providers, pharmaceutical companies, and researchers, integrating their AI platform into existing clinical workflows.

Viz.ai: Expediting Critical Care Pathways

Viz.ai specializes in AI-powered care coordination and intelligent triage for time-sensitive medical conditions, most notably stroke and pulmonary embolism. Their core product uses deep learning to analyze medical images (e.g., CT scans) for early detection of pathologies, automatically alerting specialists and facilitating faster treatment decisions. Viz.ai’s model is designed to reduce treatment delays, improve patient outcomes, and optimize hospital resource allocation. Their commercial approach focuses on direct sales to hospitals and health systems, showcasing quantifiable improvements in clinical workflows and patient care metrics.

Paige AI: Revolutionizing Digital Pathology

Paige AI is a leader in computational pathology, developing AI-powered diagnostic tools for cancer. Their technology analyzes whole-slide images of tissue biopsies to assist pathologists in detecting and grading cancer more accurately and efficiently. Paige’s model aims to reduce diagnostic errors, streamline pathology workflows, and provide quantitative insights that can inform treatment. Their strategy involves integrating their AI solutions into digital pathology platforms, collaborating with academic medical centers, and pursuing regulatory clearances for their diagnostic algorithms.

Aidoc: Enhancing Radiologist Efficiency

Aidoc provides AI solutions for radiologists, designed to flag critical findings in medical images and prioritize urgent cases. Their algorithms analyze various imaging modalities (CT, MRI, X-ray) for conditions like intracranial hemorrhage, pulmonary embolism, and cervical spine fractures. Aidoc’s model enhances radiologist efficiency, reduces turnaround times for critical diagnoses, and helps prevent missed findings. Their commercial strategy involves integrating seamlessly with existing PACS systems within hospital networks, demonstrating clear ROI through improved workflow and patient management.

Butterfly Network: Democratizing Ultrasound with AI

Butterfly Network has developed a portable, single-probe ultrasound device integrated with AI capabilities. Their model aims to make ultrasound imaging more accessible and easier to use, allowing for earlier diagnosis and monitoring across various clinical settings. The AI assists in image acquisition, interpretation, and quantification, democratizing a technology traditionally confined to specialists. Butterfly’s commercial approach targets a broad range of healthcare professionals, from emergency medicine to primary care, through direct sales and partnerships.

Nabla: AI Assistant for Clinical Documentation

Nabla offers an AI assistant designed to automate clinical documentation for healthcare providers. Their technology leverages large language models to generate clinical notes in real-time during patient consultations, reducing administrative burden and allowing clinicians to focus more on patient interaction. Nabla’s model aims to improve clinician well-being, enhance documentation accuracy, and free up valuable time. Their commercial strategy involves direct-to-clinician subscriptions and integrations with electronic health record (EHR) systems.

Regulatory Posture and Market Readiness: A Critical Comparison

When evaluating these innovators, health plan executives and health IT professionals must consider not just their technological prowess, but their regulatory maturity and preparedness for global markets. The FDA SaMD Framework in the US and the EU AI Act, coupled with the EU MDR, present distinct yet increasingly interconnected compliance challenges. Viz.ai and Aidoc, operating in the diagnostic and triage space, have prioritized FDA clearances for their algorithms, demonstrating a clear understanding of their classification as SaMD. This proactive engagement with the FDA CDRH signifies a commitment to establishing clinical validity and analytical performance, which is crucial for payer adoption. For instance, Viz.ai’s numerous FDA clearances for stroke detection and pulmonary embolism triage underscore their dedication to a regulated pathway. Viz.ai FDA clearances Similarly, Aidoc has secured multiple FDA clearances for its various AI algorithms, providing a strong foundation for clinical deployment. Aidoc FDA clearances Paige AI, with its focus on primary cancer diagnosis, also exemplifies a robust regulatory strategy, having obtained FDA clearances for its pathology AI. This not only validates their technology but also builds trust with clinical users and payers. Tempus AI, while heavily involved in precision oncology and data analytics, operates in a space that can encompass both regulated SaMD and unregulated clinical decision support (CDS) tools. The distinction is critical. As I. Glenn Cohen, a leading expert on health law and bioethics, has frequently highlighted, the regulatory classification of AI in healthcare hinges on its intended use and the level of clinical decision-making autonomy it possesses. If Tempus’s AI tools provide “information to a healthcare professional to assist in rendering a diagnosis or treatment decision,” they might fall under CDS, which has a lower regulatory burden. However, if they “directly provide a diagnosis or a treatment recommendation,” they are more likely to be classified as SaMD, requiring FDA clearance. Health plans must scrutinize the specific claims and regulatory status of each Tempus AI offering. Butterfly Network, with its AI-assisted ultrasound, also navigates this dual space. While the hardware itself is a medical device, the AI features that aid in image acquisition or interpretation may be subject to SaMD regulations, depending on their intended use. Nabla, as an AI assistant for documentation, currently operates largely in the CDS realm, aiming to streamline administrative tasks rather than directly diagnose or treat. However, as these tools become more sophisticated, their potential to influence clinical decisions could bring them under increased regulatory scrutiny, particularly under the EU AI Act’s high-risk classifications. The EU AI Act, with its classification of AI systems used in medical devices as “high-risk,” introduces a new layer of complexity. The August 2026 compliance deadline for high-risk AI systems means companies must not only achieve CE Mark under EU MDR, often with the involvement of Notified Bodies like BSI Group or TUV SUED, but also adhere to the stringent requirements of the AI Act. This includes robust risk management systems, data governance, human oversight, transparency, and cybersecurity. Bakul Patel, currently Senior Director, Global Digital Health Regulatory Strategy at Google Health, and formerly an FDA digital health leader, has consistently emphasized the importance of building trust and responsible AI from the ground up, a principle that underpins both FDA and EU regulatory philosophies. Companies like Viz.ai, Aidoc, and Paige AI, with their existing QMS (ISO 13485) and regulatory clearances, are arguably better positioned to adapt to the EU AI Act’s requirements than those with less established regulatory frameworks.

The Imperative of Dual Compliance

For health plan executives and health IT professionals, the message is clear: the era of fragmented regulatory compliance is over. Companies without a defined and executed FDA SaMD pathway, coupled with a proactive strategy for EU AI Act compliance, face significant and rising risks. These include:

• Enforcement Actions: Regulatory bodies, both in the US and EU, are increasingly prepared to take action against non-compliant AI health tools.
• Health Plan Exclusion: Payers are becoming more sophisticated in their due diligence, prioritizing solutions with demonstrable regulatory clearance and robust clinical evidence. Unregulated or poorly regulated AI tools will struggle to gain reimbursement and integration into care pathways.
• Market Access Barriers: Without dual compliance, global expansion becomes a formidable, if not impossible, hurdle. The EU AI Act’s high-risk classification for healthcare AI means that companies aiming for the European market must meet these new standards by August 2026. The companies that will thrive are those that view regulatory compliance not as an afterthought, but as a core architectural principle. Their commitment to the FDA SaMD framework and proactive engagement with evolving EU regulations, including EU MDR and the EU AI Act, will be the ultimate differentiator in a rapidly maturing global market.