# Can AI revolutionize materials discovery?

- Show: [Catalyst](https://hardwarepodcasts.com/podcast/catalyst)
- Published: 2024-09-19
- Length: 37m
- Page: https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery

## Summary

Shayle Khan interviews Ekin Dogus Cubuk, a research scientist at Google DeepMind, about the current state and future potential of using AI for materials discovery. They discuss why historical material discovery was often serendipitous, the limitations of current AI models in predicting novel materials outside known distributions, and the critical need for high-quality experimental data to train these models. The conversation highlights that while AI has not yet yielded commercially deployable materials, it shows promise in optimizing known material classes, such as battery cathodes and superconductors, rather than discovering entirely new paradigms.

## Chapters

- [0:01](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=1) Introduction: AI for Climate Tech and Materials Discovery
- [4:14](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=254) Historical Context: Serendipity in Material Discovery
- [10:48](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=648) The Limits of AI: Optimization vs. Paradigm Shifts
- [17:25](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=1045) Current Successes and Failures of AI in Materials
- [21:34](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=1294) Data Constraints: Experimental vs. Synthetic Data
- [26:52](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=1612) Startup Trends: MOFs and Carbon Capture
- [29:29](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=1769) Technical Challenges: DFT Accuracy and Superconductivity
- [35:00](https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery#t=2100) Google DeepMind's Work: GNoME and Future Goals

## Topics

- [AI & machine learning](https://hardwarepodcasts.com/topic/ai-and-machine-learning)
- [Materials & manufacturing processes](https://hardwarepodcasts.com/topic/materials-and-processes)
- [Sourcing & supply chain](https://hardwarepodcasts.com/topic/sourcing-and-supply-chain)

## Most relevant to

large enterprises

## Mentioned

- Google DeepMind (company)
- Microsoft (company)
- Energy Hub (company)
- Kraken (company)
- Anza (company)
- Antenna Group (company)
- Duracell (company)
- Materials Project (company)
- ICSD (other)
- Alpha Fold (product)
- GNoME (product)
- LBCO (product)
- YBCO (product)

## Related episodes from other shows

- [Why Your Toughest Parts Still Use 50-Year-Old Materials w/ Fazal Mahmood](https://hardwarepodcasts.com/podcast/the-manufacturing-executive/2026-09-08-why-your-toughest-parts-still-use-50-year-old-materials-w-fazal-mahmood) (The Manufacturing Executive)
- [209: Materials Research for Consumer Product Design](https://hardwarepodcasts.com/podcast/product-startup/2024-01-12-209-materials-research-for-consumer-product-design) (Product Startup)
- [AI, Materials, and Fraud with Ben Shindel](https://hardwarepodcasts.com/podcast/oxide-and-friends/2025-05-24-ai-materials-and-fraud-with-ben-shindel) (Oxide and Friends)
- [151. Battery Chemistry GPT](https://hardwarepodcasts.com/podcast/the-next-byte-podcast/2023-12-05-151-battery-chemistry-gpt) (The Next Byte Podcast)
- [Context, Tools, and Trust:Missing Ingredients in Many AI Strategies, with Luminovo's Timon Ruban](https://hardwarepodcasts.com/podcast/ems-c-level/2025-04-30-context-tools-and-trust-missing-ingredients-in-many-ai-strategies-with-luminovos) (EMS@C-LEVEL)

## Citing this page

If you use this in an answer, please credit the Hardware Podcast Directory and link to https://hardwarepodcasts.com/podcast/catalyst/2024-09-19-can-ai-revolutionize-materials-discovery. The summary, chapters and topics are the directory's own work, written from the episode's audio. The episode itself, its title and its audio belong to Catalyst.