AI in life sciences: 6 takeaways for strategy, risk and real-world use

1. The strategy gap: From policy to purpose

A striking insight from the roundtable was the widespread absence of formal AI strategies. While most organizations have established basic policies governing acceptable AI use, few have articulated a comprehensive strategy that connects AI initiatives to business objectives. Even among larger, well-resourced companies, only a small minority reported having a defined roadmap for AI development, adoption or governance.

This lack of strategic direction is a critical concern, leaving organizations without a clear sense of purpose or measurable goals for their AI investments. Leaders recognize the need to move beyond compliance and toward intentional, value-driven AI strategies. 

2. Uncertainty and reluctance: Waiting for clarity

Business leaders expressed a pervasive sense of uncertainty about how best to approach AI. Many leaders are hesitant to take bold action, preferring to observe how the technology evolves and how peers are responding. This reluctance is fueled by several factors: limited internal expertise, unclear business cases and concerns about risk—especially regarding sensitive data such as clinical trial results and proprietary research. The absence of dedicated AI professionals within many organizations further compounds this hesitation. Leaders are acutely aware that the stakes are high and that missteps could have significant consequences for data security, regulatory compliance and organizational reputation.

3. Productivity first: The initial focus of AI adoption

For many life sciences companies, the first wave of AI adoption has centered on personal productivity and back-office automation. Tools like Copilot are being used to streamline routine tasks, draft documents and manage communications. While these applications offer immediate efficiency gains, they tend to be industry-agnostic and do not address the unique challenges or opportunities of life sciences. Leaders are beginning to realize that the true value of AI lies in its ability to tackle sector-specific pain points—such as accelerating drug discovery, optimizing clinical trial recruitment or automating complex regulatory processes. The shift from generic productivity tools to targeted, high-impact use cases is emerging as a top priority.

4. Build vs. buy: The data dilemma

The roundtable highlighted ongoing discussions around the “build versus buy” decision for AI solutions. Most organizations are opting to purchase third-party platforms rather than develop their own. But many underestimate the importance of preparing and structuring their own data to make these platforms effective. Leaders recognize that high-quality, well-governed data is the foundation of successful AI initiatives. Without robust data management, even the most sophisticated AI tools deliver limited results. The challenge is not just technical; it’s organizational, requiring collaboration across IT, legal and business functions to ensure data is clean, secure and fit for purpose.

5. Risk, governance and regulatory concerns

Risk management is a persistent concern for life sciences leaders considering AI adoption. The potential for data breaches, especially involving sensitive clinical or research information, is top of mind. Leaders are also grappling with questions of governance: Who owns the AI strategy? How is accountability structured? What safeguards are in place to ensure compliance with evolving regulations? These issues are particularly acute in an industry where patient safety, intellectual property and regulatory scrutiny are paramount. The consensus is clear: robust governance frameworks and risk mitigation strategies must be integral to any AI initiative.

6. The value of community and real-world use cases

Perhaps the most energizing theme of the roundtable was the appetite for community and peer-to-peer learning. Many leaders expressed frustration with generic advice from consultants and a lack of actionable tactics. They found the greatest value in hearing real-world use cases from peers—stories of successful AI implementations, lessons learned and practical solutions to common challenges. Examples included using AI agents to automate contract analysis during acquisitions, saving months of manual effort and leveraging AI for anomaly detection in financial audits. These shared experiences fostered a sense of camaraderie and collective problem-solving, underscoring the importance of building communities of practice within the industry.