
The Google I/O event took place on May 14, revealing a new version of the Android TV OS, based on Android 14.
This version is focused on reduced power consumption, and has a feature that supports picture-in-picture (PiP) mode, allowing for multitasking with one app running in full screen, and one in a small window.
The OS reduces power usage, with the low energy mode shutting off network connectivity when the TV isn’t being used.
There’s also an increased energy mode, which keeps the network active, enabling Google Home functions.
Finally, there’s an optimised energy mode, which allows vital features only to connect to network while the TV is off. This includes Google Cast (over Wi-Fi) and Google Assistant.
Google noted the PiP mode will be available on “qualified Android 14 TV models,” and that two media apps cannot be used simultaneously.
Android 14 for TV comes with a promise of improved performance and startup times (up to four seconds).
Additionally, low memory devices should get a boost in performance.
This OS version also adds new accessibility features, including colour correction for users that have colour vision deficiencies, enhanced text options for those with other vision issues, and easier navigation for those with motor impairments.
Google also revealed that Android TV OS has reached 220 million monthly active devices, a 47 per cent increase compared with 2023.
Google has also encouraged apps to be moved from old Leanback UI toolkits to the new Compose TV system.
During the Google I/O conference, AI was at the top of the mind for attendees. The new software updates packed with AI suggest the company want developers to begin creating as soon as possible.
Mobile took a back seat during the conference, and hardware wasn’t included. The Pixel 8A was revealed as a budget-friendly option for its flagship range.
Google’s AI onslaught was spearheaded by CEO Sundar Pichai, who said, “We’ve been investing in AI for more than a decade—and innovating at every layer of the stack: research, product, infrastructure, and we’re going to talk about it all today.”
“Still, we are in the early days of the AI platform shift. We see so much opportunity ahead for creators, for developers, for startups, for everyone. Helping to drive those opportunities is what our Gemini era is all about.”
The company announced at the conference that its new Gemini 1.5 Pro platform is now available globally for developers.
It will also bring its Gemini Nano platform, designed for local use, to the Chrome desktop client.
Additionally, Gemini Nano will get a multimodality upgrade for text, images, videos, and audio, designed to run on smartphones.
The company will also roll out AI overviews to millions of users through the Search Generative Experience (SGE).
A planning feature has also been added to Gemini, with Google’s Head of Search, Liz Reid saying, “Planning is an area where, whether you’re planning a trip or planning for a party, there’s typically a lot of fun involved. But there’s also a lot of work involved for that. And so we’ve been asking the question, how can we take the work out of the planning for you?”
Gemini will also replace Google Assistant as the default AI assistant, and Gemini Nano’s multimodal functions will provide users with a new smartphone experience via Android’s TalkBack feature.
This will result in deeper integration with the mobile OS and Play Store applications. Currently, it works with Gmail, YouTube, and Google Messages.
Google also demoed DeepMind’s Project Astra, a visual chatbot with spatial understanding that can accept contextual inputs.
Google Workspace services get Google 1.5 Pro as the accompanying LLM, allowing for a larger context window and other advanced features.
Gemini 1.5 Pro is now available for Workspace Labs and Gemini for Workspace Alpha users.
Google launched the Pixel 8A a week before the conference, meaning it isn’t technically a part of the event.
However, this is the first budget Pixel model with a 120Hz display refresh rate, a Tensor G3 chipset, and an Ultra HDR feature.
Developers will also gain access to a new generation of Tensor Processing Unit (TPU), and Google teased its sixth generation TPU, Trillium.
The company revealed little information about Trillium, but noted it comes with the third generation of SparseCore and offers 4.7x higher computing performance per chip compared with the fifth generation.