I just read The AI-fication of Jobs by Huy Nguyen Trieu as part of my effort to deep dive into the technical aspects of AI. I want to understand its real implications for our industry.
In pharma, AI hasn’t made the disruptive impact so many headlines suggest. Not yet. In medical devices, the progress is more tangible. Apps are collecting real-time data, supporting clinical trials, and generating insight. But in pharma, we’re still early in the adoption curve.
What the book highlights—rightly—is that AI isn’t one thing. We often speak of it as a monolith, when in reality it spans everything from faster algorithms to generative tools like ChatGPT, to true deep learning and neural networks (which still feel, in many ways, quite primitive).
What also resonated with me is the focus on data. Real, high-quality, time-sequenced, bias-checked, referenceable data. Because the most powerful forms of AI—the ones that learn—depend on that.
In highly regulated environments like life sciences, data isn’t just fuel; it’s responsibility. We need to solve for traceability, privacy, IP protection, and bias before we can trust these systems to support real scientific decisions.
So while everyone is rushing to appear “AI-ready,” I’m asking a different question: Are we data-ready?
This book doesn’t give all the answers, but it starts the right conversation.
Link to book: https://courses.cfte.education/aification/
