The key announcements from Google I/O

Google used its annual developer conference to double down on its commitment to developing new tools powered by artificial intelligence (AI).

The tech giant announced and previewed new features coming to all its major platforms, including its search engine and Chrome web browser, as well as confirming its new mixed reality headset would be released later this year.

Here is a round-up of the key announcements.

In a move chief executive Sundar Pichai called a “reimagining” of search on the web, Google confirmed it would start rolling out a version of its search engine entirely powered by its Gemini AI.

AI mode will have its own tab within search, and when in use, it will enable users to ask longer, more complex questions, as well as follow-up questions, in order to go deeper into searches more seamlessly, and in a way that would previously have taken several different searches, Google says.

It will start rolling out in the US first, starting on Tuesday, but is widely expected to be expanded to other regions in the months to come.

A striking and tangible demonstration of how generative AI can be used to help make life easier was Google’s demonstration of its new live translation tool in Google Meet, the firm’s video calling app.

Launching initially for translating between English and Spanish, the AI-powered update enables both users to speak their native language and have it instantly translated and dubbed over the call for the other to understand.

The name for Google’s plans to build AI agents, Mariner is set to launch this year with a tool that will be able to carry out tasks on the web for users autonomously.

For example, users will be able to set the agent a web-based task, such as finding a hotel in a city they are planning to visit, or a specific item they need to repair a bike, and the AI tool will be able to carry it out itself.

Google said it was still in the early stage of this technology, but had plans to eventually introduce it to its Chrome web browser, search engine and Gemini app.

At a time when debate is fierce over the impact and possible damage AI-generated media could have on the creative industries, Google has continued its push to be an industry leader in the space.

At I/O, the US tech giant unveiled an array of new tools, including its new video generation tool, Veo 3, and Imagen 4, for creating AI-generated images.

As well as adding audio generation capabilities within these, the firm also announced a new music creation tool, known as Lyria 2, and an AI-based filmmaking product called Flow.

All are capable of producing high-quality and realistic-looking media, but Google says it is working closely with the creative industries to help shape how it builds its models to work with creators, and not replace them.

Google’s plans to build new headsets and glasses with Gemini AI built in are not new, having been announced last year, but the firm did reveal its first headset, known as Project Moohan and being built with Samsung, will be released “later this year”.

In addition, it gave a first look at the first glasses made with Android XR, its operating system for headwear, which included Gemini AI built in and able to respond to queries in context based on what the user can see and hear around them.

In addition, the live translation tool unveiled for Google Meet was also shown off in the glasses, and Google confirmed fashion firms Gentle Monster and Warby Parker were on board to help build stylish frames for the technology when it comes to market.

Following in the footsteps of OpenAI, Google also announced a new hyper-premium subscription for tech early adopters who want access to all its AI advances first.

Known as Google AI Ultra, the new subscription will cost 250 US dollars (£187) a month, but will give users access to all its AI tools first, including its most powerful models.