General Motors is developing a proprietary in-vehicle artificial intelligence assistant expected to launch later this year [3].
The move represents a strategic shift toward vertical integration of software. By building its own system, GM can access telemetry, OnStar data, and predictive maintenance information that general-purpose AI models cannot reach due to structural limitations [2, 5].
While the company recently integrated Google's Gemini AI into its fleet, the proprietary system is designed to be more deeply embedded with vehicle-specific data [1, 2]. This approach allows the assistant to interact more directly with the car's hardware and internal systems than a third-party application could [5].
Integration of Google Gemini into GM vehicles is slated to begin next year [1]. This suggests a hybrid strategy where users may have access to both a broad-knowledge assistant for general queries and a specialized GM assistant for vehicle management.
Beyond the immediate launch of the AI assistant in 2026 [4], the company is eyeing more advanced automation. A hands-free, eyes-off-driving feature is set to debut by 2028 [1].
The proprietary AI will live within the in-dash infotainment systems of GM vehicles [1, 2]. The system aims to overcome the limitations of general-purpose models by utilizing the specific data streams available to the manufacturer [2, 5].
“GM is building its own in-vehicle AI assistant, slated for launch later in 2026.”
GM's decision to build a proprietary AI alongside Google's Gemini indicates that general-purpose LLMs currently lack the deep system-level access required for comprehensive vehicle control. By owning the AI layer, GM can transform the dashboard from a simple interface into a predictive tool that monitors vehicle health and integrates seamlessly with proprietary services like OnStar, reducing reliance on external tech giants for core automotive functions.
