AI in Clinical Trials Projected to Reach $8 Billion by 2030 as Industry Unveils New Tools

Analysts have projected that AI in clinical trials will represent an $8 billion business segment by 2030, generating up to $110 billion per year in value to pharma. The forecast comes as the industry confronts what one pharmaceutical executive describes as a "clinical insight latency" problem—the gap between when something happens in a trial and when researchers have enough information to act on it.

At least 88% of organizations already have a business function using AI, although only about 23% are deploying agentic AI—meaning a lot of bots who are orchestrated together to do something important for an organization. Only 5% of companies are reporting that all the AI being embedded into their systems has had any material value to them.

The SCOPE event in Orlando experienced record-breaking attendance exceeding 4,800 people, with multiple companies announcing new AI-powered platforms and capabilities. Trialbee released the initial suite of AI-powered capabilities within their Honey Platform, turning real-world recruitment data into actionable intelligence. The initial suite of capabilities is highlighted by AI-generated candidate summaries, which bring key eligibility information to the forefront and significantly reduce the time research sites spend reviewing referred patients' history and data for faster processing. Other capabilities include duplicate patient and spam detection, automatic masking of potential PII, intelligent AI chatbot, and patient access optimization.

ConcertAI unveiled Accelerated Clinical Trials (ACT), an enterprise agentic AI platform designed to automate and inject predictive intelligence into the overall study process. It integrates real-world and proprietary data with advanced AI workflows to help sponsors and contract research organizations (CROs) shorten trial timelines by 10 to 20 months and dramatically reduce costs. ACT is built on CARAai, the company's multimodal agentic AI platform, and deploys a suite of purpose-built assistants and agents to automate critical trial activities such as literature reviews, protocol design, competitive trial analysis, feasibility assessments, site selection, and patient matching. Development teams can use its design and writing tools to slash design timelines and costly protocol amendments by 50%. The platform's automated validation strategies can also reduce timelines related to site selection, activation, and recruitment by 25% to 50%.

PhaseV launched its AI-powered Enrollment Lab solution, an additional layer to its ClinOps platform, enabling sponsors to quantify a study's enrollment potential and model the impact of constraints and trade-offs prior to site identification. The company's population-first approach looks to accelerate traditional site-level surveys with electronic health records data for real-time modeling of enrollment dynamics. Study teams can use Enrollment Lab to explore alternatives and evaluate how specific inclusion and exclusion criteria impact patient volume.

Barcelona-based AI company Biorce announced it has closed a $52.5 million Series A round. The financing includes new investment from DST Global Partners, with existing investors Norrsken VC and YZR Capital increasing their participation, alongside participation from Mustard Seed Maze and Endeavor Catalyst. The mission of Biorce is to make clinical trials faster, as well as more reliable and accessible, on a global scale. Its Aika platform is built on a data foundation of 1 million clinical trials and designed to anticipate risks, reduce errors, and eliminate protocol amendments to accelerate the development of new therapies by up to 50%.

WCG unveiled ClinSphere Trial IntelX, its next-generation predictive intelligence solution powered by more than 80,000 complete protocols and 40,000 operationally benchmarked trials, to aid sponsors and CROs in the planning, designing, and execution of clinical trials. Key features of the tool include agentic AI for enrollment and performance forecasting, scoring of participant and site burden, enrollment forecasting and operational risk alerts supporting adaptive methodologies, explainable AI coupled with expert review, and a portfolio optimization module. Syneos Health was revealed as the first customer to adopt Trial IntelX.

Medable announced the launch of its third agentic AI agent, this time for helping research sites reduce burden and assisting principal investigators in the oversight and monitoring of electronic clinical outcome assessment (eCOA) data. The company previously released agents for automating trial master file processes and clinical trial monitoring.

The head of IT globally for development operations at Bristol Myers Squibb emphasized that realizing AI's potential requires redesigning workflows rather than bolting AI onto existing processes. Due to the exponential rate of change that has been occurring with modern AI, key aspirations for what clinical operations will look like inside of five years revolve around four pillars, the first of which is "autonomous clinical workflows." How work gets done will no longer be linear, manual, or reactive, and will involve planning multi-step workflows, executing across systems, continuously monitoring outcomes, and escalating only when human judgement is required.

Among the key technical shifts making this feasible are autonomous agents, agentic architecture, and agentic AI. The old tenet to "develop a process first, then design the technology to support it" needs to be completely abandoned. The value will be realized not by bolting on AI to existing longstanding processes, which are functional but highly inefficient.

Multimodal AI—the kind that can read documents, images, and video—will be a key enabler, freeing humans of the work of acquiring, digesting, and synthesizing all that information. Neuro-symbolic AI will likewise be important, combining the ability to recognize something with rules to justify a recommendation or insight, and could therefore support decisions on a regulatory pathway.