You won't need to say 'Hey Google' anymore.
At CES 2025, Google announced the replacement of its aging Assistant with Gemini on Google TV — a shift that reframes the television not as a passive screen but as a conversational presence in the home. The move reflects a broader cultural moment in which the boundary between human intention and machine response continues to dissolve, one living room at a time. Whether this deepens our relationship with technology or simply makes it harder to escape remains the quiet question beneath the announcement.
- Google is retiring the familiar 'Hey Google' trigger entirely, betting that people are ready to speak to their televisions the way they speak to each other.
- Gemini's ability to parse intent — not just words — marks a genuine leap: asking for 'kid-friendly movies like Jurassic Park' now yields results that actually understand what you meant.
- Far-field microphones embedded in new TV hardware mean the room itself becomes the interface, removing the remote as a required intermediary for voice interaction.
- Ambient sensors will detect your approach and surface personalized information — weather, news, calendar — before you've said a word, blurring the line between appliance and attendant.
- The television is being repositioned from a device you activate to one that activates around you, raising both the promise of seamless convenience and the unresolved tension of always-on awareness in the home.
At CES 2025, Google unveiled a sweeping overhaul of Google TV, replacing the older Google Assistant with Gemini — a far more capable AI that understands conversational speech, context, and intent. The change is immediately felt in how users interact with their screens. There's no longer a need to say 'Hey Google.' Instead, you speak naturally, and Gemini listens. In one demonstration, a user asked for movies like Jurassic Park but appropriate for children, and the system surfaced fitting options without hesitation.
The capabilities extend well beyond entertainment. Users can ask Gemini to explain the solar system in terms a third grader would understand, or to help plan a beach vacation — and the answers appear on screen alongside relevant YouTube videos, honoring the visual nature of the medium. Voice-generated AI screensavers and smarter smart home controls round out the feature set.
On the hardware side, Google is working with manufacturers to embed far-field microphones directly into television sets, allowing conversation from across the room. More ambitiously, new ambient sensors will detect when someone approaches the TV and automatically display personalized widgets — weather, news, calendar entries — without any prompt. When no one is in the room, the television can power itself down entirely.
Taken together, these features sketch a larger vision: a television that is less a device you switch on and more a presence that adapts to the rhythms of your day. Google's direction is clear, even if the deeper implications — of a screen that watches back, anticipates, and never quite sleeps — are still unfolding.
Google is about to make talking to your television feel like talking to another person in the room. At CES 2025, the company unveiled a significant overhaul to Google TV powered by Gemini, its AI assistant that has been gaining ground in the broader AI race. The shift replaces the older Google Assistant with something considerably more capable—a voice interface that understands conversational speech, context, and nuance in ways the previous system did not.
The most immediate change is how you'll interact with your screen. You won't need to say "Hey Google" anymore. Instead, you can speak naturally, the way you'd ask a friend for help. In a demonstration, a user asked to see movies like Jurassic Park but suitable for children. Gemini understood the request immediately and surfaced appropriate options. That kind of contextual understanding—grasping not just the words but the intent behind them—is the core of what Gemini brings to the living room.
Beyond entertainment queries, the system handles far more complex requests. A user asked Gemini to explain the solar system in terms a third grader would understand, and in another instance, to help plan a vacation to a beach destination. Both requests generated answers displayed directly on the television, accompanied by related YouTube videos. This matters because television is a visual medium; text-based web results don't serve the same purpose as video content does. The system also lets you generate AI-powered screensavers with your voice and control smart home devices more intuitively through the TV itself.
The hardware side of the upgrade is equally significant. Google is working with manufacturers to build far-field microphones into upcoming television sets, which means you can converse with your TV from across the room without shouting into a remote. But the more ambitious feature is what Google calls an ambient experience. New televisions will include sensors that detect when you're approaching the screen. When triggered, the TV displays personalized widgets—weather forecasts, news stories tailored to your interests, calendar information—without you having to ask. The same sensors can display art when you're nearby but not actively watching, or power the television down entirely when the room is empty.
These features point toward a larger vision: making the television less of a device you turn on to watch something and more of an integrated part of your home's daily rhythm. It's a shift from passive consumption to ambient intelligence, where your screen anticipates your needs and adapts to your presence. Whether this actually changes how people use their televisions, or whether it simply adds another layer of always-on technology to the home, remains to be seen. But Google's bet is clear: the future of TV is conversational, contextual, and aware.
Notable Quotes
You'll now be able to speak to it more seamlessly, using natural language and conversing with Google TV without needing to say 'Hey Google'— Google's CES 2025 demonstration
The Hearth Conversation Another angle on the story
Why does Google need to replace Google Assistant entirely? Couldn't they just add Gemini features on top of what already exists?
The old Assistant was built for command-and-control—you say a specific thing, it does a specific thing. Gemini is built for conversation. It understands context, can follow a thread of thought, can handle ambiguity. Bolting it on top of the old system would create friction. A clean replacement lets the TV think more like a person.
The ambient sensing feature—displaying widgets when you walk by—that sounds like it could get annoying fast. What stops it from becoming visual clutter?
That's the real design challenge, isn't it? The idea is that the widgets are personalized and contextual. Your weather and calendar, not random ads. But you're right to be skeptical. It depends entirely on execution. If it's thoughtful, it's useful. If it's not, it's just another screen demanding attention.
Does this require a new TV, or will existing Google TVs get these features?
The software upgrades—the Gemini voice capabilities—those can roll out to existing devices. But the hardware stuff, the far-field mics and the ambient sensors, those require new sets. So you get a choice: upgrade your software now, or wait for new hardware if you want the full experience.
What's the privacy angle here? A TV with sensors that know when you're in the room?
That's the unspoken question in all of this. Google hasn't detailed how the sensor data is handled, whether it's processed locally or sent to the cloud, what gets stored. For a device in your living room, that matters. People should know what they're agreeing to.