According to Mirit Eldor, managing director of Life Sciences Solutions at Elsevier, the use of AI in pharmaceuticals has three dimensions: the data used to train AI models; the technology itself; and the expertise associated with it. Elsevier worked alongside the Pistoia Alliance to address AI data challenges, such as synergy between ontologies and taxonomies, and to ensure AI has the right context to give the right answers . With AI technology encompassing extractive, predictive and generative models, the technology requires preparation and maturity to be the right solution to the right problem.
Another AI challenge is the place of humans in the loop. With this in mind, we spoke with Eldor about the challenges of AI and how collaborative approaches help.
Most people are excited about AI, but don’t know how to use it. Elsevier and the Pistoia Alliance have conducted extensive market research to determine where customers want AI, if they are comfortable with AI, and how we can help them. Most people (between 95% and 97% of our market research respondents) believe that AI can help them work faster, better and more efficiently; they believe that AI can free up time to work on higher value-added projects.
However, many people were also concerned. The main concerns were about AI leading to misinformation and poor decisions, including critical errors. Our challenge is to help alleviate these concerns. There’s no doubt that AI can be great, but we need to use the technology better. The challenge is how to ensure that the technology works well enough that we can rely on it. That’s what I see from this research: people want to make sure that they can rely on AI so that when they go into a clinical trial, they can make the best decisions about what to do. what to do and how to design the test.
We started using AI over a decade ago. So we had time to understand the risks and how to mitigate them. At the time, there was no legislation, so we had to develop a set of responsible AI principles that could be applied to every algorithm and AI-based product we build. There are five main elements: the real impact of the solutions; measures to prevent the creation or reinforcement of unfair prejudices; be able to explain how our solutions work; creating accountability through human oversight; and respect privacy and defend data governance. The latter includes the protection of intellectual property, which is particularly important in the life sciences. We make sure we have clean, validated data, the right technology, and a knowledgeable human.
Any data we use to train algorithms or provide answers must pass through legal governance and controls. People know us primarily as a publisher, but we have extensive datasets in the life sciences domain. We pay royalties for it and make sure we have the right to use it. We apply strict controls over data management and confidentiality.
Knowledge sharing is really important for working with other stakeholders and like-minded people. Whether we work with pharmaceutical innovators, service providers, governments or NGOs, we all face similar challenges and can learn from each other. In 2022, I was invited to testify before the U.S. Chamber of Commerce, which was examining AI governance and how to mitigate its risks. I am always happy to work with regulators, industry bodies and government organizations. Pistoia is a great example. It’s a pre-competitive collaboration of stakeholders and organizations that are grappling with these questions around data, how to bring together data sets and how to ensure that we have data standards. It’s a perfect community for sharing and learning from others.
Good question! I think we need to continue to think about how to use AI responsibly. In science, we think about patterns so that there is no false information. We must continue to ensure data integrity, and we must remain attentive and careful when deploying AI to make critical decisions.
AI will be widely used to optimize preclinical processes. People currently use it for simple tasks to achieve better productivity, but AI can go much further. I’m really excited. Elsevier has been around since 1840. We have survived by being able to change and adapt over time to many of the technological disruptions that everyone was wary of. We have understood how to use these technologies for our benefit, from the advent of CD-ROM and the Internet to today with AI. It’s natural to be uncertain, but technology always makes life easier and more efficient. Yes, we’re still figuring it out, but I’m very excited about the difference AI is going to make.