Biotech · 2 min read

175 AI Drug Programs Have Entered Trials but Zero FDA Approvals So Far

Roughly 175 AI originated drug programs have entered human trials but zero hold full FDA approval as of August 2026, putting the verification era ahead of the hype cycle.

By Classy AI News · August 17, 2026

175 AI Drug Programs Have Entered Trials but Zero FDA Approvals So Far

Zero approvals, 175 programs

As of August 2026, no AI discovered drug holds full FDA approval, according to industry trackers and published analyses, despite roughly 175 AI originated programs that have entered human trials and an estimated $60 billion invested in AI drug discovery since 2019.

That gap between pipeline volume and regulatory proof is the story biotech cannot ignore.

Where the pipeline actually sits

A peer reviewed analysis presented at ASCO in 2026 counted 117 AI enabled therapeutic assets across 63 companies in interventional human trials. Of those, 60 (51.3%) had completed Phase 1 and only 8 (6.8%) had completed Phase 2. FDA approval records do not classify products by whether AI was used in discovery, so field wide first approval claims depend on each tracker's methodology.

Insilico Medicine's rentosertib program for idiopathic pulmonary fibrosis reached Phase III in July 2026, with both biological target and chemical structure identified by generative AI, according to the company's announcement. No AI originated drug has received regulatory approval yet.

Isomorphic Labs, the DeepMind spinout, raised $2.1 billion in May 2026 and confirmed multiple proprietary candidates in late preclinical development, with CEO Demis Hassabis targeting first in human trials by end of 2026. At Davos in January 2026, Hassabis acknowledged a slip from the previously stated end of 2025 target.

Regulatory framework arriving

On January 14, 2026, the FDA and European Medicines Agency jointly published the Guiding Principles of Good AI Practice in Drug Development: ten high level principles spanning nonclinical research, clinical trials, manufacturing, and post market safety monitoring.

The principles cover human oversight, risk based management, data governance, model design, lifecycle management, and clear context of use. For teams building AI native discovery stacks, auditability is no longer optional.

What investors should watch

Phase II success rates for AI discovered molecules sit near traditional industry averages according to published analyses. Capital efficiency at the company level remains unproven: Recursion cut staff, BenevolentAI delisted, and the verification era has begun with Phase III rentosertib rather than an approval headline.

Sources

  1. Insilico Medicine, Rentosertib Phase III announcement (July 7, 2026)

Newsletter

Get the dispatch

One field. One email when we publish. Privacy.