General Motors is bringing Google Gemini-powered AI assistants to its cars, trucks and SUVs starting in 2026, the automaker announced at its GM Forward event in New York. The move puts GM squarely in competition with Mercedes' ChatGPT integration and Tesla's xAI Grok rollout as automakers race to deploy conversational AI that can understand natural speech patterns and provide contextual responses.
General Motors just fired the latest shot in the automotive AI wars. The Detroit giant announced Wednesday that it's bringing Google Gemini-powered conversational AI to its entire fleet by 2026, setting up a direct confrontation with Mercedes' ChatGPT integration and Tesla's xAI Grok deployment.
The announcement came during GM's Forward event in New York, where executives positioned the Gemini assistant as the next evolution of their existing Google partnership. GM vehicles already run Google's built-in operating system, giving drivers access to Google Assistant, Maps, and apps directly through the infotainment screen. But this new integration goes much deeper.
"One of the challenges with current voice assistants is that if you've used them, you've probably been frustrated by them because they're trained on certain code words or they don't understand accents very well," Dave Richardson, GM's senior VP of software and services, told TechCrunch. "What's great about large language models is they don't seem to be affected by that."
The timing couldn't be more strategic. While GM's more ambitious tech overhauls - including hands-off, eyes-off autonomous driving - won't arrive until 2028, the Gemini assistant will start rolling out next year via over-the-air updates to OnStar-equipped vehicles from model year 2015 and above. That's potentially millions of GM vehicles getting an AI upgrade through the Play Store.
Richardson's team has been quietly building toward this moment. In 2023, Google began using its Dialogflow chatbot to handle non-emergency OnStar queries like routing and navigation. The new Gemini integration promises to handle more complex requests - drafting messages, planning multi-stop routes with charging stations, even answering questions like "What's the history of this bridge I'm driving over?"
But GM isn't putting all its eggs in Google's basket. Richardson revealed that the company plans to test foundational models from OpenAI, Anthropic, and other AI firms as part of its broader strategy to develop custom AI trained specifically on vehicle specifications and user data.
