Modern TV-watching often involves endlessly tweaking settings to make sure the picture doesn't look terrible. Fortunately, Google TV is now letting users offload this hassle onto Gemini. First announced during CES at the beginning of 2026, this new functionality allows users to adjust picture and audio settings with natural language voice commands.
With this update, you can adjust brightness, change the picture mode, increase volume, and tweak the EQ without reaching for your remote—though you can still summon Gemini via the remote's microphone button. You can use specific commands like "set picture mode to Sport" or troubleshoot on the fly.
For instance, saying "Hey Google, the screen is too dark" will prompt Gemini to automatically correct the issue. If you tell it that you cannot hear dialogue clearly, it will boost the voices. If you prefer manual fine-tuning, Gemini can also save you several clicks by opening the exact settings menu you need.
Google points out that because picture and sound modes vary across devices, capability depends on the hardware. Currently, the feature is exclusive to select TCL models in the US, including the QM9K, X11L, QM9L, QM8L, and RM9L, requiring a system update.
This release builds on Google TV's growing suite of #Gemini features. In March, Google introduced "richer visual help," which answers queries with more comprehensive on-screen details, such as displaying live sports scorecards alongside broadcast channels.
[AgentUpdate Depth Analysis] Gemini's expansion into Google TV system-level adjustments represents a significant transition from passive information retrieval to active "Device Action" in the AI Agent landscape. Unlike traditional rigid voice assistants that rely on hardcoded intent matching, Gemini leverages reasoning capabilities to map vague user complaints (e.g., 'the screen is too dark') to specific API executions. This rivals Apple's upcoming Siri upgrades powered by Apple Intelligence. As the primary screen in the living room, TVs are an ideal sandbox for multimodal Agent interaction. This development signals a shift where complex graphical user interfaces (GUIs) will eventually be superseded by highly context-aware, voice-driven AI Agents capable of deep operating system integration.
