# Redwood Materials found a very practical use for AI: Sorting old batteries

> Source: <https://www.machinebrief.com/news/redwood-materials-found-a-very-practical-use-for-ai-sorting-0ysw>
> Published: 2026-11-21 15:28:01+00:00

**Key takeaway:** Redwood Materials trained a custom model on its own battery data and runs it on an on-site GPU cluster, combining camera, sensor, label, shape and weight inputs to sort batteries at over 98% accuracy and several tons per hour.

# Redwood Materials found a very practical use for AI: Sorting old batteries

[Business Insider](https://www.businessinsider.com)

Tesla cofounder JB Straubel's Redwood Materials says its AI-powered sorting line identifies old batteries with more than 98% accuracy.

- A version of this story originally appeared in the AI Insider newsletter.
- Sign up for Business Insider's weekly [AI Insider newsletter here](https://www.businessinsider.com/subscription/newsletter/tech-memo) .

[Recycling batteries](https://www.businessinsider.com/powering-the-ai-data-boom-with-recycled-ev-batteries-2026-11) sounds like a great idea. Until you realize someone has to sort through mountains of them first.

It's a tricky job. Batteries come in all shapes and sizes, with different [chemical ingredients](https://www.businessinsider.com/redwood-materials-jb-straubel-evs-batteries-new-black-gold-tesla-2025-4). Mixing up the wrong ones can make recycling less efficient and more expensive.

Enter [AI](https://www.businessinsider.com/redwood-materials-recycled-ev-batteries-power-ai-data-centers-2026-10).

[Redwood Materials](https://www.redwoodmaterials.com/news/how-computer-vision-is-recycling-batteries-at-industrial-scale/), the battery recycling startup founded by Tesla cofounder JB Straubel, has built an automated sorting system that can identify batteries with more than 98% accuracy, according to the company.

Here's how it works.

Old batteries arrive in a jumble of boxes and bins. Machines separate them into a single-file line. Each battery then passes through a tunnel equipped with cameras and sensors that examine its label, shape, and [weight](https://www.machinebrief.com/glossary/weight).

AI figures out what kind of battery it is and sends it to the right place for recycling. Redwood developed an AI model for this, trained on the company's voluminous battery data, and it runs on a cluster of GPUs on site.

The system was built in just a few months and can handle several tons of batteries an hour.

Why does this matter?

Redwood wants to recover valuable materials such as lithium, nickel, and cobalt, rather than relying so heavily on mining and imports.

This is a neat example of AI tackling a messy, unglamorous industrial problem. And potentially making American manufacturing more competitive.

*Sign up for the AI Insider newsletter* [here](https://www.businessinsider.com/subscription/newsletter/tech-memo)*. Reach out to me via email at* [abarr@businessinsider.com](mailto:abarr@businessinsider.com)

[Business Insider](https://www.businessinsider.com/redwood-materials-ai-battery-sorting-recycling-2026-11)

Get AI news in your inbox

Daily digest of what matters in AI.
