Megan Baca moderated Ropes & Gray’s annual “From the Boardroom” panel – held in San Francisco during the 2024 J.P. Morgan Healthcare Conference – which this year looked at the role of artificial intelligence and big data in the context of dealmaking. It can feel hard to escape AI at the moment, with some debate as to whether AI is currently over-hyped or in fact at a transformational tipping point.
The panel of experts raised a number of points that will be of interest throughout 2024.
- Building a moat: A key point for any would be founder in AI is to demonstrate what is their company’s “moat”. What value is unique about the company and can this be protected from copying by competitors. This could include protected proprietary intellectual property, a superior customer experience or a dominant role in being used by others in the sector. Without this protection investors can be more nervous to deploy capital in a company.
- Importance of business models: Relatedly, many AI companies start with the technology and only think about the business model later. However, investors want to see a clear understanding of expected sales and sources of revenue. If this will include big pharma, companies should start discussions with them early to assess what technologies (and features) will actually be important to them. For example, big pharma and financial VCs will want to understand the source of data and whether the appropriate patient consents are in place.
- Bespoke deal terms: AI transactions are developing unique deal terms, which can be tailored to the specific technology. For example, data monetisation is often an important area of negotiation. Terms can include a question of whose data is it, how it can be used and what payments that prompts.
- Public versus proprietary data: AI relies on large data sets to become as powerful and relevant as possible. Many AI companies have been built on public data sets. While these can be valuable, they are not the complete picture – often missing negative results. Big pharma will then want to run AI models on their own proprietary data. However, there is a nervousness in combining that proprietary data with public data, meaning big pharma may need to develop their own internal AI functionality. Therefore, collaboration remains even more important in an AI / big data context, although this can be challenging from a competition perspective.
- Role of regulators: Healthcare regulators are continuing to try and understand the role and application of AI. However, those in the industry need to engage with regulators early and often, perhaps providing them with white papers and mode guidance than they would in other areas. Regulators are understandably focused on patient safety, the protection of patient data, as well as ensuring that AI is used responsibly. AI companies need to be able to demonstrate that they can effectively make the link between the AI models and what actually goes on in the lab / clinical trials.
- Role of patients: Patients themselves have an important role in all of these discussions. To be effective, AI requires broad data sets that reflect the diversity of patient populations, which is particularly important in ensuring that there are no biases in the results. Patients need to be brought on the AI journey, which may include explaining that the same animal and human clinical trials have been undertaken (even if some AI models can limit the need for these). Patients will likely want more control of their data, perhaps including how it is monetised.
Whatever your view, it is clear that many more companies will be trying to be demonstrate how they use AI in 2024. There continue to be many more opportunities, but to be successful in the long-term, companies need to be focused on the business case, regulatory compliance and the role of patients.