AI Transforms Raw Data into Valuable Insights, Just Like an Oil Refinery
At the 2025 Bioprocessing Summit held in Boston, a panel of industry experts gathered to discuss the role of Artificial Intelligence (AI) and Machine Learning (ML) in the biopharmaceutical industry. The summit saw a divide within the industry regarding the adoption of digital transformation and AI, with some parts needing more convincing while others are already embracing it.
Mark Mackey, Chief Security Officer (CSO) of Cresset, expressed concerns about the potential use of large language models (LLMs) and synthetic data to generate "missing data" in the industry. He emphasised the importance of ensuring that AI tools are used responsibly and ethically.
Irene Rombel, CEO of Biocurie, was also a panelist at the summit. She suggested that the focus should shift from gathering as much data as possible to getting the right data. Rombel emphasised the importance of subject matter expertise and robust statistical analysis, arguing that quality trumps quantity when it comes to data.
Colin Zick, managing partner at Foley Hoag, and Cenk Undey, global iCMC digital transformation program lead at Sanofi, were also part of the panel. Zick highlighted that AI is a tool for critical thinking, not a replacement for it. He advised reading the FDA's new draft guidance on how AI could support regulatory decision making.
Undey expressed surprise at the focus on topics like explainable AI and risk-based applications in Good Manufacturing Practice (GMP), and expressed a desire for more discussion on AI's robotics and GenAI frontiers as applied to bioprocessing. He emphasised the importance of minimising prediction errors by addressing the learning aspect of AI/ML models as they evolve and adapt.
One common pitfall in the use of AI/ML tools is jumping right into them without thinking about the fundamentals, such as basic statistics, understanding the data, and the problems being solved. Undey advised minimising prediction errors by addressing the learning aspect of AI/ML models as they evolve and adapt, emphasising the importance of risk assessment, point of use, and documenting the computation method.
Data is intrinsically dirty and requires investment in time and money to extract any value, with AI tools acting as refiners of the data. The panelists agreed that in today's technology-driven era, data is an incredibly valuable raw material.
Undey stressed that using AI is not the objective, but using it where it makes sense and asking the right questions to solve the right problems is more critical. Sometimes an AI/ML method is not suitable for a task, but a simple ordinary process might do the job better. Developing and advancing data products for decision making is a critical step in the product development and biopharma manufacturing continuum, according to Undey.
Lori Ellis, editor of BioSpace Insights, commented on the use of generative artificial intelligence at the summit. She highlighted that it is a common myth that ML/DL models need millions of data points. The quality of the data is more important than the quantity, she said.
In conclusion, the panelists agreed that while AI holds great potential for the biopharmaceutical industry, it should be used thoughtfully and responsibly. Zick summed up his approach toward AI by saying that it should make everything better and should never make anything worse.
