ArcaScience Launched Flow to Automate Drug Assessments
The platform uses specialized AI models to extract adverse-reaction data from biomedical literature for regulatory filings.
Updated on Sept. 23, 2026 in Biotech
Live Poll
Do you trust AI software to accurately manage safety-critical data in pharmaceutical drug development?
ArcaScience has launched Flow, an AI software system designed to generate regulatory-grade drug benefit-risk assessments. The platform integrates 100 billion biomedical data points to track emerging safety evidence.
Why it matters
The system streamlines the labor-intensive process of monitoring and reporting drug safety profiles as new scientific evidence emerges. This capability helps pharmaceutical companies maintain regulatory compliance by automating the synthesis of complex clinical literature.
Flow utilizes a multi-model architecture featuring 24 specialized AI engines to process 100 billion biomedical data points. In a benchmark test using 600 adalimumab abstracts, the system achieved a 96.3% recall rate by identifying 104 out of 108 defined adverse-reaction concepts.
The players
ArcaScience
A research-focused company that builds AI architectures for pharmaceutical data analysis and drug safety monitoring.
Sanofi
A global pharmaceutical company focused on immunology and vaccine development that partners with ArcaScience.
AstraZeneca
A multinational biopharmaceutical company that leverages ArcaScience platforms for clinical data research.
GSK
A global healthcare firm utilizing ArcaScience technology to support pharmaceutical research and safety workflows.
Takeda
A research-driven pharmaceutical company that integrates ArcaScience software into its drug development pipeline.
The details
Flow functions by structuring evidence across diverse clinical areas while maintaining source-level traceability for all findings. To identify risks, the system analyzes text from biomedical abstracts to recover and link specific adverse-reaction concepts—medical terms denoting side effects or negative patient outcomes—to established drug profiles. This allows the AI to provide a granular audit trail for regulatory-grade submissions that would otherwise require manual synthesis.
Timeline
September 23, 2025: ArcaScience raised $7 million in a seed round.
September 23, 2026: ArcaScience launched the Flow AI software.
The Tech Race
The launch of Flow follows the growth trajectory established by the firm’s $7 million seed round raised in September 2025. This deployment marks a shift from early research development to an operational tool now being integrated by 20 pharmaceutical partners.
Pharmaceutical researchers and compliance teams can now utilize Flow to automate the collection of safety data from vast literature sets. The software is currently being deployed across international markets, including the US and UK, to support ongoing regulatory requirements.
The takeaway
Flow demonstrates a high degree of precision in automating drug safety literature reviews, outperforming traditional manual screening. Researchers should monitor whether this model performance remains consistent as the company expands its integration with the 20 existing pharmaceutical partnerships.
Further reading
For more on the current landscape of automated evidence synthesis, visit Biotech.
Source note: This article includes information reported by Firstwordpharma.
Live Poll
Do you trust AI software to accurately manage safety-critical data in pharmaceutical drug development?
