Having studied medicine, I’ve always been fascinated by the intersection between clinical practice and technology. My medical background gave me a deep appreciation for how communication, accuracy, and time management directly influence patient outcomes. When I began exploring generative AI tools, I quickly realized how transformative they could be for healthcare if applied responsibly.
My first encounters with AI in healthcare were experimental: testing language models, generating patient summaries, and exploring how speech recognition could capture clinical conversations. It quickly became clear to me that generative AI could do more than just automate tasks; it could support the way healthcare professionals think, document, and communicate. Imagine having an assistant that can structure notes, translate conversations in real time, or help summarize a complex consultation without losing nuance, these are not distant possibilities anymore. It even made me start my own company: Wellcom Health.
At the same time, my experience in this business has made me acutely aware of the risks associated with relying on AI in healthcare. Unlike other industries, mistakes here can have direct consequences for people’s health. Generative models are impressive but not infallible; they can misinterpret or fabricate information. That is why I believe that accuracy, transparency, and human supervision must remain at the core of every AI-driven healthcare solution.
Exploring frameworks like Retrieval-Augmented Generation (RAG) has shown me a promising way forward. By grounding AI outputs in verified medical data, we can significantly reduce the risk of misinformation while still benefiting from the model’s generative capabilities. This balance between innovation and reliability will define the success of AI in healthcare.
Ultimately, my journey with generative AI has reinforced a belief I already held from my medical studies: technology should never replace human judgment or empathy. Instead, it should empower healthcare professionals to work more efficiently, communicate more effectively, and focus more deeply on what matters most, the patient.
Great reflection and interesting to hear from someone with a medical background. What I like about your part is that you emphasize accuracy, transparency and human supervison. This ties back directly to the EU’s AI act.
Would be interesting to know from your perspective, would the EU AI act, slow adoption too much or are they needed to ensure public trust?
The EU AI-act is a good thing for Europe, we would get flooded with US “Cowboy” companies that have much more money and innovation power than we have. Now they don’t even bother entering our HealthTech market.
However, this “protection” definitely slows down innovation. But the same can be said about Medical Device Regulation, ISO13485 en ISO27001 certification in this field. In the US you only need HIPAA approval and off you go!
Great question. Let me know if you have more.