General Motors is teaming up with Google to bring artificial intelligence into millions of vehicles, and the transformation is already underway. GM will add a conversational AI assistant powered by Google Gemini to its cars, trucks, and SUVs starting in 2026, marking a significant shift in how drivers interact with their vehicles.
The partnership represents more than just a voice assistant upgrade. GM’s integration with Gemini builds on the automaker’s existing relationship with Google, which already powers the infotainment systems in Buick, Chevrolet, Cadillac, and GMC vehicles with Google Maps and Google Assistant.
This move puts GM in an escalating automotive AI competition alongside rivals like Stellantis, Mercedes, and Tesla, each racing to deliver more natural and capable in-car experiences. The technology promises to understand context, remember previous conversations, and respond to drivers in ways that current voice assistants simply can’t match.
How the GM and Google Gemini In-Car AI Assistant Works
The Gemini-powered AI assistant operates through a combination of advanced large language models, existing Google infrastructure, and GM’s OnStar platform to deliver conversational capabilities that access vehicle systems and external information.
Conversational AI: Natural Communication and User Experience
The system leverages large language models to understand natural speech patterns without requiring specific commands or keywords. Unlike traditional voice assistants that struggle with accents or variations in phrasing, the Gemini assistant uses context from previous conversations to interpret requests more accurately.
Drivers can speak naturally when asking questions or giving commands. The conversational AI assistant handles tasks like drafting messages, planning multi-stop routes that include charging stations or coffee shops, and answering questions about surroundings like the history of a bridge they’re driving over.
The generative AI approach means the assistant adapts to how individual drivers communicate rather than forcing them to learn specific voice commands. This makes interactions feel more like talking to a passenger than operating a device.
Integration With OnStar and Google Built-In
GM vehicles from Buick, Chevrolet, Cadillac, and GMC already run Google Built-In operating systems that provide access to Google Assistant, Google Maps, and apps through the infotainment screen. The Gemini integration builds on this existing foundation rather than replacing it entirely.
OnStar serves as the backbone for the in-car concierge system. GM previously used Google Cloud’s Dialogflow chatbot for non-emergency OnStar features like routing assistance, and the Gemini assistant represents an evolution of that capability.
The assistant connects to vehicle systems through OnStar to access real-time data and control functions. GM plans to take base models and train them on specific vehicle specifications, then run those refined models directly on the computing platform.
Vehicle Data and Personalized Features
The assistant taps into vehicle data to provide maintenance alerts and explain car features like one-pedal driving. It can identify potential issues and report them directly to drivers based on sensor readings and diagnostic information.
Key personalized capabilities include:
- Pre-conditioning climate control before drivers enter the vehicle
- Learning driving habits to offer customized route suggestions
- Providing maintenance recommendations based on actual vehicle usage
- Accessing individual preferences for frequently visited locations
Drivers control what information the assistant can access and use. GM emphasized that any collected data goes toward product improvement rather than being sold for additional revenue, with opt-in and opt-out controls built into the system.
Over-the-Air Upgrades and Compatibility
The Gemini assistant will arrive as an over-the-air upgrade available through the Play Store for OnStar-equipped vehicles. This delivery method means drivers won’t need to visit dealerships for installation.
The upgrade reaches back to model year 2015 and above vehicles with OnStar. This compatibility gives the technology access to millions of existing GM vehicles already on the road rather than limiting it to only new purchases.
While GM is launching with Google Gemini, the automaker plans to test several foundational models from other AI firms including OpenAI and Anthropic. This multi-vendor approach suggests GM wants flexibility in choosing the best technology rather than committing exclusively to one provider.
What This AI Assistant Means for the Future of Driving
GM’s partnership with Google signals a shift toward vehicles that handle more driving tasks independently while raising questions about data ownership and positioning the automaker against rivals racing to dominate automotive AI.
Hands-Off and Autonomous Driving Capabilities
At the GM Forward media event, General Motors announced plans to introduce eyes-off driving by 2028, debuting on the 2028 Cadillac Escalade IQ. This represents a step beyond the current Super Cruise system, which already covers 600,000 miles of hands-free roads across North America.
The conversational AI assistant will help bridge the gap between today’s driver assistance and tomorrow’s autonomous driving. It can explain features like one-pedal driving to new electric vehicle owners and provide real-time maintenance alerts by accessing vehicle data.
GM’s approach combines Super Cruise’s track record—700 million miles driven without a single reported crash attributed to the system—with over five million fully driverless miles from Cruise technology. The 2028 computing platform will deliver 35 times more AI performance for autonomy, giving vehicles the processing power needed for advanced self-driving features.
The Escalade IQ will serve as the testing ground for this next generation of personal autonomy, uniting propulsion, steering, infotainment, and safety systems on a single computing core.
Privacy, Security, and Data Control
GM’s AI assistant will require driver permission to access vehicle intelligence and personal preferences through OnStar. This permission-based approach puts control in the hands of owners, but it also raises questions about what data gets shared with Google and how it’s stored.
The system will eventually transition from Google Gemini to GM’s own custom-built AI. This shift suggests General Motors wants to reduce dependence on third-party platforms and maintain control over the data flowing through its vehicles.
Drivers will need to weigh the convenience of personalized route suggestions and predictive maintenance against the amount of vehicle and behavioral data being collected. The assistant’s ability to spot maintenance issues early and recommend dinner destinations requires continuous monitoring of driving patterns, location history, and vehicle performance metrics.
Competition in the Automotive AI Space
General Motors faces competition from tech companies like OpenAI, Anthropic, Mistral, and xAI that are all developing AI systems for various applications. While these companies haven’t focused specifically on automotive AI, their language models could power rival in-car assistants.
The automotive AI space is heating up as traditional automakers partner with tech giants to avoid falling behind. GM’s choice of Google Gemini positions it alongside Android Automotive, though the company previously removed support for Android Auto and Apple CarPlay from some vehicles.
Other manufacturers are pursuing different strategies—some developing proprietary systems, others licensing technology from multiple AI providers. GM’s dual approach of launching with Google Gemini before transitioning to its own AI gives it flexibility to compete regardless of which direction the market moves.
The new computing platform works with both electric vehicles and gas-powered models, allowing GM to scale the technology across its entire lineup rather than limiting it to premium EVs. This broad deployment could give General Motors an edge in collecting real-world data to train future AI systems.
More from Steel Horse Rides: