MedTech companies risk creating more problems than they solve if they deploy artificial intelligence on fragmented data and inconsistent processes, according to discussions at the Veeva MedTech Summit in Amsterdam.
The conference focused less on how quickly the sector can adopt AI than on whether organizations have the foundations needed to use it reliably. Across presentations, panels and interviews, a recurring theme was that reliable data, standardized processes, suitable governance and accountable people must be in place first.
Medtronic’s four-year preparation
Becki Brasher of Medtronic described the company’s four-year effort to become ready to implement AI. Instead of starting with the technology, Medtronic worked on organizational groundwork: harmonizing processes, improving data consistency, consolidating systems and setting up governance structures.
Brasher said Medtronic did not begin with AI; it had to earn the right to use it by first building the necessary foundations. The implication was that AI introduced into fragmented environments can produce outputs that are difficult to trust, validate or scale.
Medtronic’s experience also pointed to workforce readiness. Technology alone cannot deliver transformation, and employees need knowledge, confidence and appropriate oversight to use new tools in their daily work, according to the discussion.
Governance embedded from the start
In the closing keynote, Erik Orzechowski and Benjamin Rochette addressed the tension between moving quickly and maintaining quality, safety and regulatory standards. They argued that governance should be built into technology development from the outset rather than treated as a final compliance step.
That requires closer collaboration between IT, quality and regulatory teams. IT understands implementation, while quality and regulatory professionals assess risks and determine the evidence needed to show compliance, the speakers said.
They also called for proportionate governance. An AI application supporting low-risk administrative work may not need the same controls as one affecting patient safety or regulatory decisions. Clear responsibilities, early risk assessments and suitable controls can give organizations a more flexible route to innovation while preserving obligations.
Human judgment and measurable value
The summit also emphasized that AI should assist rather than replace human reviewers. Brasher said human judgment will remain essential until organizations build trust that what AI agents deliver is accurate. AI can analyze information, generate documentation and make recommendations, but its output still needs scrutiny, particularly in a regulated environment.
Speakers compared relying on AI without understanding its subject matter to using a calculator without knowing the mathematics behind it. For the next generation of regulatory and quality professionals, the concern is that over-reliance could leave fewer people able to challenge inaccurate recommendations.
On business value, the closing keynote identified time savings, reduced process cycle times, less rework, improved quality and employee adoption as measures. Organizations were urged to resist pursuing every opportunity at once and to start with controlled pilots that retain human oversight and build confidence before scaling.
Orzechowski coined the phrase “slow down to speed up” to capture the point. The broader conclusion was that MedTech’s AI progress will depend less on the number of initiatives launched than on the ability to turn promising applications into reliable, scalable capabilities.