Robin AI System Autonomously Discovers Drug Candidate for Blindness in Lab-Confirmed Breakthrough — First End-to-End Autonomous Scientific Discovery

The Information Technology and Innovation Foundation published a policy analysis on June 2, 2026, of Robin — a multi-agent AI system developed by FutureHouse that completed the first documented end-to-end autonomous scientific discovery: hypothesis generation, experimental validation, and laboratory confirmation, all without human direction at any step. Robin autonomously identified ripasudil — a glaucoma treatment drug — as a repurposing candidate for dry age-related macular degeneration (dAMD), the leading cause of blindness in the developed world, a connection that had never previously appeared in scientific literature. Robin analyzed 551 relevant scientific papers in approximately 30 minutes (versus an estimated 540 hours of manual expert review), generated the therapeutic hypothesis, and triggered laboratory experiments that confirmed the drug's efficacy in preclinical models. The underlying study was published in Nature. The ITIF analysis framed Robin as a policy inflection point: if AI can autonomously conduct the full scientific discovery loop, drug repurposing programs — particularly for neglected tropical diseases where commercial incentives are weak — could be systematically automated at low cost. BCG data cited in the paper found 5 of 15 AI-assisted drug candidates reached clinical trials in under 4 years, versus the historical 5–6 year average.