Why AI Won't Save the 90% of Clinical Trials That Still Fail
Why AI Won't Save the 90% of Clinical Trials That Still Fail  
Podcast: BioSpace
Published On: Thu Aug 21 2025
Description: While AI represents a significant advancement in efficiency for early-stage drug discovery, it won't dramatically change the 90% clinical trial failure rate. Most failures stem from fundamental gaps in biological understanding rather then the processes where AI is able to have the most impact.The discussion highlights AI's strengths and ability to reduce preclinical costs. However, they caution that AI faces significant limitations in predicting complex biological properties like toxicity due to insufficient data, and regulatory acceptance of AI-only safety assessments remains unlikely, meaning traditional clinical trials will continue to be necessary.This episode is presented in partnership with ⁠Cresset⁠.Host⁠⁠⁠⁠⁠⁠⁠⁠Lori Ellis⁠⁠⁠⁠⁠⁠⁠⁠, Head of Insights, BioSpaceGuests⁠Mutlu Dogruel⁠, VP of AI, Cresset⁠Mark Mackey⁠, CSO, CressetDisclaimer: The views expressed in this discussion by guests are their own and do not represent those of their organizations.