😺 🎙️ Watch: How Isomorphic Labs works to drug "undruggable" diseases

2026-05-06 · Source: The Neuron · Field: Health & Wellbeing — Pharmaceuticals & Biotechnology, Artificial Intelligence & Machine Learning, Life Sciences & Biology · Depth: Intermediate, long

Summary

The Neuron's May 06, 2026, brief highlights an interview with Isomorphic Labs, a Google DeepMind spinout, on their AI-driven drug discovery efforts. Isomorphic Labs, founded by Demis Hassabis, is applying advanced AI, including technology derived from AlphaFold 3, to tackle "undruggable" diseases. Their new system reportedly more than doubles AlphaFold 3's accuracy in critical drug design stages. The podcast features Becky Paul (medicinal drug design) and Michael Schaarschmidt (foundational AI research) discussing the challenges of traditional drug discovery, where new drugs take over a decade and roughly $6B to market with a 90% clinical trial failure rate. They detail how AI is accelerating structure prediction, enabling the design of novel molecules, and exploring concepts like molecular glues for protein degradation. The brief also includes other AI-related interviews covering spatial intelligence, AI music generation, AI agents for real-world tasks, and robotics in manufacturing.

Key takeaway

For pharmaceutical investors evaluating R&D pipelines, recognize that AI-first drug discovery, exemplified by Isomorphic Labs, significantly de-risks and accelerates the drug development timeline. Your investment strategies should account for companies integrating advanced AI like AlphaFold, as they are poised to bring novel therapies to market faster and more efficiently, potentially transforming the landscape of "undruggable" diseases like KRAS-mutated cancers.

Key insights

AI, particularly AlphaFold-derived technology, is dramatically accelerating drug discovery and tackling previously "undruggable" targets.

Principles

• AI can replace slow experimental steps in drug design.
• Molecular glues offer a new paradigm for drug action.
• AI models can make inexplicable leaps in discovery.

Method

Isomorphic Labs uses an AI-first approach, leveraging advanced structure prediction models to design novel molecules and accelerate drug candidate identification, aiming for one design cycle to yield a drug candidate.

In practice

• Explore AI for protein structure prediction to reduce experimental validation.
• Investigate molecular glue mechanisms for new therapeutic avenues.
• Consider AI for tasks requiring complex molecular design.

Topics

• Isomorphic Labs
• AI Drug Discovery
• AlphaFold
• Protein Folding
• Medicinal Chemistry

Best for: Investor, AI Scientist, Research Scientist, Director of AI/ML

Related on AIssential

• Why We're Not Close to AI Drugs Yet | Isomorphic Labs — Drug design comprises multiple distinct ML problems, not a single one, despite AlphaFold 2's protein structure prediction success.
• ‘An AlphaFold 4’ – scientists marvel at DeepMind drug spin-off’s exclusive new AI — Isomorphic Labs' proprietary IsoDDE model significantly advances AI drug discovery, outperforming open-source alternatives in binding affinity prediction.
• DeepMind spinout Isomorphic Labs raises $2.1bn — AI-driven drug discovery benefits from specialized teams and distinct corporate structures.
• 🗞️ Techie uses ChatGPT and AlphaFold to build DIY mRNA cancer vaccine, saves dog — AI tools empower individuals without specialized training to conduct complex scientific research and develop novel solutions.
• AlphaFold: Grand Challenge to Nobel Prize with John Jumper — AlphaFold's success demonstrates AI's profound utility in accelerating scientific discovery, even without full interpretability.
• The Future Of Drug Discovery Is 4 Billion Years Old (Viswa Colluru, Founder & CEO at Enveda) — AI-driven exploration of nature's chemistry offers a potent, overlooked pathway for discovering first-in-class, affordable medicines.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Neuron.