Venture capital is shifting focus from AI drug discovery to the expensive, slow-moving machinery of clinical trials.

Designing a new molecule in a digital lab takes weeks, but testing it in humans still takes years. This mismatch is the quiet crisis of modern biotechnology. While hundreds of startups focus on finding new drugs, the actual bottleneck remains the clinical trial phase, where over 90 percent of candidates ultimately fail.

Faro’s recent $37 million Series B funding highlights a growing industry realization. Discovering a compound is useless if the trial design is too flawed to prove it works. The influx of capital suggests that investors are growing weary of AI-designed drugs stalling in regulatory limbo. The bottleneck is no longer the lab; it is the clinic.

The Protocol Problem

Traditional clinical trials rely on document-heavy protocols that are slow to write and difficult to adapt. Faro and competitors like QuantHealth, which recently secured a $45 million Series B, are replacing these static documents with structured, AI-driven data models.

These platforms simulate trials and optimize protocols before the first patient is even recruited. Major drugmakers are paying attention. Faro has already secured partnerships with Bristol Myers Squibb and Recursion to scale these autonomous systems.

The Reality Check

Can software truly predict human biology? While AI can optimize patient cohorts and streamline operations, it cannot eliminate the biological uncertainty of clinical testing. The real test for these platforms is not just raising capital or signing pilot programs. It is proving they can actually lower that stubborn 90 percent failure rate in real-world settings.