First-of-its-kind solution automates validation of Tables, Listings, and Figures (TLFs) in clinical trials, marking a major step forward toward more automated statistical programming for life sciences.

Beaconcure, a clinical biometrics technology company, today announced the commercial launch of Verify Intelligence, the independent AI review module of its Verify platform.

Every regulatory submission depends on hundreds or thousands of tables, listings, and figures (TLFs). Most sponsors and CROs check these outputs with double programming plus manual review. Both methods rely on people working from the same specification, so they tend to share the same blind spots. When an error gets past both, it can reach a regulator.

Verify Intelligence adds a check that works differently. It reads the study specification and source data, produces its own version of each output, and compares it with the version the programming team delivered. When the two disagree, it shows the reviewer where the difference is and traces it back to the source so it can be resolved. Each step is recorded, so reviewers can see how every result was reached. The AI flags the discrepancies and the statisticians and programmers decide.

“Double programming has been the industry’s safety net for decades, but two people reading the same specification can make the same mistake,” said Seth Houston, CEO of Beaconcure. “Verify Intelligence gives biometrics teams a second reviewer that approaches the work from a different direction. It shows its work, so the experts who sign off on a submission can see exactly why something was flagged.”

Verify Intelligence was shaped by two of the world’s largest biopharmaceutical companies, whose biometrics teams defined what an independent AI review needed to do to be trusted in a regulated setting. One of those companies has now moved Verify Intelligence into production in order to protect their submissions from errors conventional validation fails to catch. The other plans to deploy in 2027.

“Going into production with a Top 10 biopharma company shows the approach holds up in a real submission environment,” said Houston. “The pilots starting next year will take it across more studies and therapeutic areas, and we expect what we learn there to shape future releases.”

The launch comes as regulators set clearer expectations for AI in drug development. In January 2026, the FDA and the European Medicines Agency jointly published ten Guiding Principles of Good AI Practice in Drug Development, which call for AI that is human-centric by design and supported by strong data governance and documentation. Verify Intelligence was designed around the same idea: AI that supports expert judgment and leaves a clear record of how each result was reached.

For sponsors that outsource statistical programming, Verify Intelligence offers an independent check on CRO-delivered outputs. Sponsors can outsource the execution without outsourcing the oversight.

Verify Intelligence works alongside the other modules of the Verify platform. Verify Translate prepares and harmonizes source data. Verify Validation centralizes TLF review with automated quality checks. Verify Control gives leaders oversight across studies. Teams can start with a single module and add others as their needs grow.

Verify Intelligence is available now to pharmaceutical, biotechnology, medical technology, and CRO organizations. To learn more or discuss a pilot, visit www.beaconcure.com.

About Beaconcure

Beaconcure is a clinical biometrics technology company. Its Verify platform protects clinical trial submissions from the errors conventional validation fails to catch, combining centralized TLF review, automated quality checks, and independent AI review in one auditable system. Beaconcure serves pharmaceutical, biotechnology, and medical device companies and CROs, and holds ISO 27001 and ISO 27018 certifications.

For more information, visit www.beaconcure.com or follow Beaconcure on LinkedIn

View source version on businesswire.com: https://www.businesswire.com/news/home/20261006261290/en/

“Double programming has been the industry’s safety net for decades, but two people reading the same specification can make the same mistake.”