According to drivers’ reports, in some cases the assistant starts speaking, then switches into listening mode and repeatedly responds to itself, creating an endless conversational loop. This bug makes the system difficult to use while driving and forces users to interact manually, which directly contradicts the purpose of the app — hands-free operation without taking attention away from the road.
Android Auto users on Reddit say the issue can occur when Gemini Live launches automatically or when there are delays in the Bluetooth connection between the phone and the vehicle. As a result, the assistant becomes “stuck” in a dialogue loop and never proceeds to execute the actual command. Another commonly reported symptom is that the assistant stops speaking after a few seconds, switches to listening mode, and then replies again instead of moving on to the next interaction — which users believe may be related to Bluetooth latency or the way audio is implemented in the vehicle’s infotainment system.
The problems are emerging at a time when Google is gradually replacing the classic Google Assistant with its new Gemini AI assistant in Android Auto, following earlier announcements that the transition is expected to be completed by the first half of 2026. While Gemini is intended to provide more natural voice interactions and better contextual understanding, its integration is still in the early stages of broad rollout, which may be affecting stability in in-car infotainment systems.
In addition to the conversational loop issue, users are also reporting degraded performance of the existing Google Assistant within Android Auto. Basic commands — such as searching for locations, controlling music playback, or placing voice calls — increasingly fail, even though the same commands work correctly outside the in-car system. As a result, some drivers have started relying on their vehicles’ native infotainment systems or resorting to manual phone operation.
Although full Gemini integration is expected to improve the comfort and naturalness of voice interactions in Android Auto — including support for more complex and context-aware commands — the current technical problems highlight the challenges of deploying large AI models in automotive environments, where safety and true hands-free operation are critical.