The story that follows is based on the quote from the extended video provided below.
In the ever-evolving landscape of technological advancements, the introduction of new technologies can often be disruptive. From the printing revolution that made scribes obsolete to the disappearance of jobs like typists, humans have consistently been central to various processes until technology took over.
At a recent event organized by consulting firm BCG during JPM, a life science executive raised an interesting question about the human role in the face of advancing artificial intelligence (AI). AI is not just coming for physical jobs but aims to replace the one capacity that sets humans apart – our ability to think and make decisions.
The field of medicine, in particular, is witnessing the rapid advancement of AI, with developers assuring that it will augment rather than replace human physicians. However, with machine learning progressing at a rapid pace, the concept of “Agentic AI” is gaining traction. This new form of AI has the ability to mimic human judgment and act autonomously without human intervention.
During the BCG event focused on digital health and AI in the healthcare industry, Thomas Senderovitz, a senior vice president of data science at Novo Nordisk, discussed the company’s efforts in building an automated clinical trials infrastructure. Their platform, FounDATA, utilizes AI algorithms to generate insights from vast amounts of clinical trial data.
Senderovitz highlighted the potential for automation in various stages of the clinical trial process, from scientific protocol design to statistical analysis and report generation. He emphasized the need for humans in the loop for now but questioned why this requirement persists when AI can perform tasks efficiently.
Novo Nordisk is actively working on automating processes and generating insights using AI, with a focus on data ethics and governance. As AI continues to transform industries, the challenge lies in ensuring trust and transparency in decision-making processes as human involvement decreases. The future may see a shift towards explainability science to track and understand AI-driven decisions. Y, cuanto menos tengamos humanos en el proceso, más decisiones no son tomadas por humanos, más necesitamos al menos rastrear y poder tener esa transparencia.”
Pero Senderovitz también reconoció un desafío dado cómo la tecnología de IA está cambiando rápidamente.