Trust and regulatory uncertainty are blocking widespread AI adoption in clinical trials, according to new data from the Pistoia Alliance, a global nonprofit focused on life sciences collaboration. Half of clinical trial professionals surveyed at the Clinical Trials Technology Congress in London cited these concerns as the biggest obstacles to implementing AI tools in drug development, even though early signs suggest AI could deliver real value to the industry.

The poll, conducted at the congress in late May 2026, paints a picture of an industry caught between opportunity and caution. While 42% of respondents reported seeing early signs of return on investment from AI, and another 23% expect ROI in the near future, the path forward remains clouded by questions about how to validate, audit, and explain AI decisions to regulators and sponsors.

What's Blocking AI Adoption in Drug Trials?

The barriers to AI adoption in clinical development are surprisingly straightforward. Respondents identified several key obstacles that are slowing the rollout of AI tools across the industry:

• Trust Gaps: Clinical trial professionals worry about the reliability and transparency of AI systems, particularly when patient safety is at stake.
• Regulatory Uncertainty: Unclear guidance on how regulators will evaluate and approve AI-driven approaches leaves companies uncertain about compliance requirements.
• Black-Box Models: AI systems that cannot explain their decisions create anxiety for both sponsors and regulatory agencies reviewing trial data.

Dr. Becky Upton, President of the Pistoia Alliance, emphasized that the solution lies not in avoiding regulation but in partnering with it. "Regulators are not the enemy," she stated. "They can be partners to pharma, working together to formalize AI guidance that supports both sides."

Becky Upton, President of the Pistoia Alliance

"A recurring message at the Congress was that regulators are ready to embrace AI, and are keen for pharma companies to engage early to ensure adoption is safe and compliant. The panel rightly emphasized that speed without control is not enough when patient safety is at stake. For AI to support clinical development at scale, the industry needs validated, auditable and explainable approaches, not black-box models that create uncertainty for sponsors and regulators alike," said Dr. Becky Upton.

Dr. Becky Upton, President of the Pistoia Alliance

This message came directly from regulators at the congress, including representatives from the MHRA (Medicines and Healthcare products Regulatory Agency), the Danish Medicines Agency, and the Swedish Medical Products Agency. Rather than resisting AI, these agencies signaled they are ready to work with pharmaceutical companies on implementation, provided the approaches are transparent and scientifically sound.

Where Can AI Actually Help Clinical Trials?

Despite the trust concerns, the poll revealed where clinical trial professionals believe AI will have the most immediate impact over the next three to five years. The top opportunities are concentrated in two areas: data management and patient engagement.

• Data Cleaning and Analysis: 48% of respondents cited data cleaning, data analysis, and insight generation as the areas where AI will have the most impact, addressing one of the most time-consuming and error-prone aspects of clinical trials.
• Patient Sourcing and Engagement: 22% identified sourcing and engaging patients as a key opportunity, reflecting a persistent industry challenge as more AI-discovered drug candidates compete for limited trial sites and patient populations.
• Real-World Data Integration: 60% of respondents are already using, piloting, or exploring patient-generated data to inform clinical development decisions beyond traditional marketing applications.

Zahid Tharia, Director of Open Pharma Research and organizer of the Clinical Trials Technology Congress, noted that the influx of AI-discovered drug candidates is creating new pressures on the system. "This influx also means more programs are competing for the same trial sites and patient populations," he explained. "The poll shows clinical development professionals believe AI can help address some of these pressures, particularly in sourcing and engaging patients, which remains one of the industry's most persistent challenges."

How to Build Trust in AI for Drug Development

Moving forward, the industry and regulators are focusing on concrete steps to address the trust gap. The Pistoia Alliance is convening pre-competitive working groups that bring together pharmaceutical companies, technology providers, and regulators to develop common frameworks and best practices.

• Early Regulatory Engagement: Pharma companies should initiate conversations with regulators early in AI implementation, rather than waiting until a drug candidate is ready for approval, to ensure compliance and build confidence in the approach.
• Validated and Auditable Systems: AI tools used in clinical trials must be validated against known standards and produce auditable decision trails that regulators and sponsors can review and understand.
• Ethical Data Collection Standards: As social media listening and patient-generated data become more common in clinical development, establishing standardized, ethical frameworks for collecting and using this data is essential.

The Pistoia Alliance has already begun developing best-practice frameworks for the ethical use of social media data in clinical development. Thierry Escudier, Clinical Portfolio Lead at the Alliance, noted that "the next step is ensuring these data are being collected in an ethical and standardized way." The Alliance is calling on pharmaceutical companies to fund and suggest new projects in the clinical space to expand this work.

Thierry Escudier, Clinical Portfolio Lead at the Alliance

The broader message from the congress is clear: AI is not a threat to clinical development, but its adoption requires a shift in how the industry approaches regulation. Rather than viewing regulators as obstacles, pharmaceutical companies are being encouraged to see them as partners in building systems that are both innovative and trustworthy. As more AI-discovered drug candidates move through the pipeline, this collaborative approach may become essential to keeping the pace of drug development moving forward without compromising patient safety.