For the first time, a drug candidate created entirely by artificial intelligence has successfully advanced to phase 2 clinical trials, marking a major milestone in the race to shorten drug development from a decade to just over a year. The experimental treatment for pulmonary fibrosis, a progressive lung disease, was designed using generative AI and reached human trials in under 18 months, a fraction of the typical timeline that often spans ten years and costs more than $2 billion.
The breakthrough was achieved by Insilico Medicine, a biotechnology company that built its entire discovery process around AI. Its platform, Pharma.AI, includes a tool called Chemistry42, which generated and evaluated 78,000 virtual molecules before narrowing the list to 60 candidates most likely to succeed. From that shortlist, researchers selected the lead molecule that is now being tested in patients. The results, published in a peer-reviewed medical journal in 2025, represent the first clinical validation of a drug designed end-to-end by AI.
To move beyond a single success, Insilico has expanded its platform with additional tools. PandaOmics processes scientific literature and trial data, while a newer addition, PandaClaw, introduces agentic capabilities that allow the system to run longer and more complex experiments with minimal human oversight. This combination enables the company to tackle multiple diseases simultaneously, rather than focusing on one drug at a time.
Scaling AI Drug Discovery Across the Industry
Recognizing that no single company can develop enough treatments alone, Insilico has licensed its platform to other pharmaceutical firms. As of now, 13 of the world’s top 20 pharmaceutical companies are using the system to run their own drug discovery projects. These collaborations are targeting life-threatening conditions including cancer, heart disease, and degenerative neurological disorders.
This strategy shifts Insilico from being just another drug developer into an infrastructure provider, similar to how tech giants power entire industries with their cloud services. By selling access to its AI tools, the company enables broader research while positioning itself at the center of a new drug discovery ecosystem. For patients, this could mean faster access to treatments for diseases that currently have few options.
The next phase will be watching how these licensed partnerships perform in clinical settings. If the platform continues to produce viable drug candidates at scale, it could redefine how the pharmaceutical industry approaches research and development. The hope is that AI-driven discovery becomes standard practice, turning a slow and costly process into one that is rapid, repeatable, and ultimately more responsive to patient needs.