AI Can Now Design Drugs in Seconds; We Still Can't Tell You If They Work (philippdubach.com)

Isomorphic Labs, the Google DeepMind spinoff, has unveiled its revolutionary Drug Design Engine (IsoDDE), a unified in silico system that dramatically accelerates computational drug discovery. IsoDDE integrates protein structure prediction, ligand binding, affinity estimation, and pocket identification, enabling it to generate drug candidates in seconds, significantly outpacing traditional methods that took days. In benchmark tests, IsoDDE achieved a 50% success rate for complex molecular predictions, far surpassing AlphaFold 3 and conventional techniques. Despite this significant leap in computational efficiency, the challenge remains: no AI-designed drug has received FDA approval yet, highlighting the gap in translating these advancements to effective treatments. The implications for the AI/ML community are profound, as Isomorphic has secured partnerships with major pharmaceutical firms like Eli Lilly and Novartis, valued at over $4 billion, although cautious milestone structures suggest these companies are awaiting clinical validation before fully committing. While IsoDDE demonstrates promise in improving drug design processes, it has yet to address the efficacy challenges that persist into Phase II trials, where success rates for AI-discovered drugs mirror those of traditional methods. Nonetheless, the potential for IsoDDE to tackle these challenges—especially through its advanced modeling capabilities—could pave the way for increased R&D productivity in a sector grappling with high costs and lengthy development timelines. The industry will be closely watching as Isomorphic progresses towards clinical trials, targeted for late 2026, to see if these computational advancements yield tangible medical breakthroughs.