A new study tackles a burning question: how well do drugs discovered by artificial intelligence fare in clinical trials? With AI making waves in drug discovery, this question is becoming increasingly relevant.

The challenge lies in defining what constitutes an “AI-discovered drug.” AI is rapidly transforming the field, leading to a surge in AI-driven drug and vaccine development. However, concerns linger regarding their success in clinical trials.

To address this, researchers conducted a first-of-its-kind analysis of clinical pipelines from companies solely focused on AI-powered biotechnology. Their findings are intriguing.

In Phase I trials, AI-discovered molecules boast an impressive 80-90% success rate, significantly exceeding historical industry averages. This suggests that AI excels at designing or identifying molecules with characteristics ideal for drug development.

However, the picture isn’t entirely rosy. Phase II trials show a success rate around 40%, albeit based on a limited sample size. This aligns more closely with historical averages.

Despite this, the study highlights promising signs for the clinical potential of AI-discovered drugs. While Phase II success rates require further investigation with larger studies, the initial results are encouraging. AI’s ability to identify promising drug candidates could significantly accelerate the development of new treatments.

This article was originally published on iBIO NewsBrief. Gain a head start on your day with iBIO NewsBrief. Subscribe to receive top industry headlines delivered straight to your inbox.