Lowe's quietly deployed 50+ AI models before the hype cycle hit

Long before ChatGPT scrambled every corporate roadmap, Lowe’s was already elbow-deep in machine learning. The Fortune 50 home improvement giant now runs more than 50 models across pricing, forecasting, and supply chain, according to Chandhu Nair, the company’s SVP of Data, AI, and Innovation. That’s not a typo. While most retailers are still figuring out where to stick a chatbot, Lowe’s has been at this for nearly a decade.

Nair, in a conversation published by OpenAI, frames AI not as a shiny object but as a natural extension of a business built on solving complex, personal problems. “Unlike most retail experiences, shopping for home improvement is very different,” he said. A customer doesn’t just need a faucet—they’re facing a leak, a renovation, and a crisis of confidence. The company’s bet is that AI bridges the gap between a warehouse full of SKUs and a homeowner paralyzed in aisle 12.

That vision materialized in two tools built directly with OpenAI: Mylow, a consumer-facing assistant designed to mimic a knowledgeable associate, and Mylow Companion, which gives actual store employees instant access to project steps, specs, and inventory. The pitch is pragmatic—democratize the expertise locked in a 100-year-old company’s institutional memory. No fluff about AGI, just a utility knife for decision-making.

But the real story isn’t the tools. It’s the operational scaffolding Lowe’s built when the hype hit. A governance committee. A community of practice. A framework tied to how customers shop, how the company sells, and how people work. Nair’s team had already been tinkering with transformer models alongside OpenAI before ChatGPT was a household name. When the board suddenly started asking questions in late 2022, they weren’t scrambling. They were ready. That’s the rare, unsexy part of enterprise AI that actually determines whether the technology sticks or becomes another pilot-program graveyard.