MedTech organisations risk creating more problems than they solve if they introduce artificial intelligence into fragmented data and inconsistent processes, according to discussions at the Veeva MedTech Summit in Amsterdam. Across presentations, panel sessions and interviews, speakers argued that successful digital transformation depends on getting the fundamentals in place first: reliable data, consistent processes, appropriate governance and clear ownership of both.
Medtronic spent four years preparing for AI
Becki Brasher of Medtronic described a four-year programme to become ready to implement AI. Rather than starting with the technology, the company harmonised processes, improved data consistency, consolidated systems and established governance structures. Brasher summarised the approach in her session: "We didn't start with AI. We had to earn AI."
Her account illustrates why deploying AI in fragmented or inconsistent environments can produce outputs that are difficult to trust, validate or scale. Medtronic also stressed preparing employees, who need the knowledge, confidence and oversight to fold new tools into everyday responsibilities; technology alone, the company argued, does not deliver transformation.
Governance built in, not bolted on
The summit's closing keynote, given by Erik Orzechowski and Benjamin Rochette, focused on balancing innovation with regulatory compliance. As adoption accelerates, they said, MedTech organisations must find ways to move quickly without compromising the quality, safety and regulatory standards the industry is held to. Their central argument was that governance should be embedded in technology development from the outset rather than treated as a final compliance exercise.
That requires closer collaboration between IT, Quality and Regulatory teams: IT understands the technology and its implementation, while quality and regulatory professionals assess risk and define the evidence needed to demonstrate compliance. The speakers also called for proportionate governance, noting that an AI application supporting relatively low-risk administrative work does not require the same level of control as one influencing patient safety or regulatory decisions. Clear responsibilities, early risk assessments and appropriate controls, they argued, allow a more flexible approach, with compliance and innovation treated as complementary rather than opposing priorities.
Human judgement stays in the loop
Despite improving capabilities, the summit emphasised retaining human judgement and professional expertise. Brasher said AI should assist human reviewers rather than replace them, and that human judgement will remain in place until organisations build trust that what agents deliver is accurate. AI can analyse large volumes of information, generate documentation and make recommendations, but its outputs still require scrutiny — a convincing answer is not necessarily a correct one, particularly in a highly regulated environment.
The closing keynote compared relying on AI without understanding its underlying subject matter to using a calculator without knowing the mathematics behind it. The comparison raises a workforce question: if employees lean on AI before building the necessary underlying knowledge, organisations could end up with fewer people able to challenge inaccurate or inappropriate recommendations. The consensus was that AI should complement human expertise, with accountable individuals retaining responsibility for important decisions.
Measuring whether AI delivers value
With more potential applications available, organisations were urged to resist pursuing every opportunity at once and instead target existing operational challenges with clear measures of success. The closing keynote listed several ways to assess the value of AI:
- time savings;
- reduced process cycle times;
- less rework;
- improved quality;
- employee adoption.
Starting with controlled pilot projects was recommended to identify where AI is most effective, retain appropriate human oversight and build confidence before expanding what works. Because the technology and regulatory expectations will continue to evolve, governance structures need to adapt alongside new capabilities. Success, the summit suggested, will depend less on the number of AI initiatives an organisation launches than on its ability to turn promising applications into reliable, scalable business capabilities.
Orzechowski's phrase for the overall message was that organisations sometimes need to "slow down to speed up". The summit's conclusion was optimistic but pragmatic: progress requires closer collaboration within companies, as well as continued engagement between manufacturers, technology providers, regulators and other industry stakeholders.