As the healthcare industry steadily explores the potential of AI (Artificial Intelligence), clinical trials stand between both innovation and caution.
During a crowded and dynamic panel session at the 2025 ARCS Conference, leaders, including Maree Beare, CEO of Clinials, gathered to examine how AI is truly being applied in trial settings, what barriers remain, and how we can move forward responsibly.
The conversation (AI in Action: Bridging Innovation and Practicality in Clinical Trial) was more than theoretical. It was a pragmatic, feet-on-the-ground, exploration of how AI is reshaping real-world processes, and how human judgement remains at the center of it all.
Framing AI in the Clinical Trial Context
While mainstream discourse often focuses on AI for drug discovery or diagnostics, the panel brought to light lesser-known, but equally impactful, use cases: AI as a tool for content generation, document simplification, and team enablement during early-stage trial setup.
“We have a unique new use case of AI in clinical trials. It’s not just analytics anymore. It’s about real-time support for documentation and communication.”
Maree described how her team uses AI to accelerate the production of plain language summaries and other operational documents; not replacing clinical professionals, but saving time and improving clarity for both patients and practitioners.
Governance, Trust, and Responsibility
One of the most candid parts of the discussion centered on governance: how to ensure that AI outputs are trustworthy, compliant, and transparent. With trial teams operating across jurisdictions, this is far from simple.
“If we have to be in Australia, here are the local laws. If we’re in the EU or the US, here’s what that means. The governance piece is never just one-size-fits-all.”
Speakers emphasized the need for shared accountability, combining the expertise of clinicians, legal teams, data scientists, and operational staff to establish AI oversight mechanisms. The audience agreed: without robust frameworks, trust would remain elusive.
“Can we use AI to help teams comply? That’s the opportunity. But if we just add AI for the sake of it, that’s where it breaks down.”
Bridging Innovation with Practicality
Innovation is only valuable if it delivers results; and here, the panel delivered some home truths. AI implementation isn’t about hype; it’s about creating tangible, operational value.
“What we’re wanting to get to in the end isn’t just automation. It’s confidence across the team, the sites, the sponsors.”
Maree (Clinials) shared examples where Clinials’ AI cut down protocol review times and improved stakeholder understanding. It’s proving especially useful in diverse, cross-functional teams where not everyone speaks the same technical language.
“We know how hard it is to actually get through a protocol. AI should make that easier, not more complicated.”
There was consensus that small, targeted deployments of AI, rather than massive system overhauls, are often the most effective path forward.
Human-Centric AI
A major point of emphasis was the need to keep AI human-centered. Technology should support, not sideline, the professionals who ensure the ethical and scientific rigor of clinical research.
“I want to be me plus the super smart machine. Not replaced, but empowered.”
Panellists advocated for an augmentation model: one where AI takes care of repeatable, high-volume tasks, giving professionals more space to focus on judgement, nuance, and communication.
“I just want to jump in on an example that really shows how much time we’re saving for teams but also how much clarity we’re giving to patients...”
Cultural Readiness and Adoption Barriers
Despite the positive use cases, there was no glossing over the realities: AI adoption in clinical trials still faces significant headwinds.
Fear of job displacement, resistance to change, confusion about what AI can (and can’t) do, and a lack of internal alignment between IT and clinical teams were all named as friction points. Regulation, while catching up, remains a hurdle in many jurisdictions.
“A couple years ago, this felt like science fiction. Now, it’s a question of how fast we can train, test, and deploy.”
Rather than recommending sweeping digital transformation, the panel called for incremental experimentation: pilot small projects, prove their value, and expand from there, with buy-in from both leadership and operational teams.
The Power of Narrative
Perhaps the most unexpected insight from the session was the importance of storytelling in driving AI adoption. Technical explanations alone don’t change minds. Factual, and clear, stories, use cases, and plain language do.
“And if anyone can explain to me how it [a power test] actually works without a stats degree, I’d love to hear it. That’s what we’re up against. AI can help close that gap.”
Maree noted that many people fear AI not because of the technology itself, but because they don’t understand how it fits into their work. By clarifying its role, and grounding it in real benefits, organizations can actually drive broader acceptance. And reap the benefits.
Conclusion: Augmentation, Not Disruption
The panel didn’t promise revolution, and that’s precisely why it resonated.
AI in clinical trials is not a magic wand but it is a powerful tool when used with clarity, purpose, and care. The way forward is not to rush adoption, but to ground it in governance, usability, and respect for the humans at the centre of the process.
“We’re not here to replace. We’re here to support. And if AI can give back time to teams and clarity to patients, then that’s where the real magic happens.”
