Google I/O 2026: the biggest announcements for creatives

Google I/O 2026 brought Adobe integrations, new AI video tools, smarter search, and the closest thing to genuine smart glasses the company has yet shown. Here's what matters for video and creative professionals.

Google's annual developer conference, this year I/O 2026, took place yesterday and featured the ubiquitous avalanche of AI announcements. Some of it was genuinely interesting, some of it was Google proving it can do things other people have already done. Here's what caught our attention.

Adobe adds creativity connector

Adobe is expanding its partnership with Google to bring the Adobe for creativity connector to Gemini in the coming weeks. Already available in the likes of Claude, it means that Gemini users will soon be able to simply describe what they want to create and have Adobe’s pro-grade tools across imaging, design, and video handle the orchestration behind the scenes.

Also, Adobe Premiere mobile is coming to Android. "With Premiere, you’ll get access to exclusive templates and effects to create and post YouTube Shorts from the app, helping you stand out in the feed," said Google.

Ask YouTube

Ask YouTube is a new conversational search feature that lets users ask more complex queries, such as "find creator reviews of cozy games to play before bedtime" rather than a keyword string. It pulls from across YouTube's full catalogue, long-form and Shorts, and serves an interactive structured response with follow-up questioning built in. It's currently available to Premium members aged 18 and up in the US via youtube.com/new, with a broader rollout to follow.

Gemini 3.5 Flash

Gemini 3.5 Flash is a faster model designed for agentic tasks, better coding capabilities, and the new generation of interactive web UIs. It is now the default model in the Gemini app and AI Mode in Search. Gemini 3.5 Pro follows next month. Alongside the model update, Google is rolling out a redesigned Gemini app it calls "neural expressive," which turns out to mean new animations, a new font, and haptic feedback.

Gemini Omni: video from anything

The more interesting model announcement for creatives is Gemini Omni. Unlike Veo, which focused on text-to-video generation, Gemini Omni is a multimodal AI model capable of generating editable videos from combinations of text, images, audio, and existing video clips. It also comes with improved real-world physics simulation.

Future versions could generate images from sound, audio from video, or in an interestingly circular model, fully interactive generated experiences themselves. 

The first version, Gemini Omni Flash, is rolling out globally to AI Plus, Pro, and Ultra subscribers, with free access arriving on YouTube Shorts and the YouTube Create app this week. API access is expected within weeks. All content generated with Gemini Omni carries SynthID watermarking. Clips are currently capped at 10 seconds.

Google Flow improvements 

Google's AI filmmaking tool Flow has been quietly evolving since its launch last year. Earlier this year, Google merged three previously separate products (Flow, Whisk (the visual mood board tool), and ImageFX (text-to-image)) into a single unified workspace, consolidating image generation, video creation, and editing into one interface. With Nano Banana (formerly ImageFX) now built into the core experience, users can create images and immediately use them as ingredients and frames for Veo video generations without leaving the app.

Gemini Omni Flash is now also accessible inside Flow for paying subscribers, which means the platform is rapidly evolving into what Google hopes is a single-stop shop for AI-assisted production work. Flow TV, an in-app showcase, lets users see the exact prompts and techniques behind clips they find interesting, which is an interesting (and for the current industry, unusual) practical learning resource.

AI detection comes to Chrome and Search

Google is expanding its AI detection tools to Chrome and Search, using SynthID watermarking and C2PA Content Credentials to reveal whether images have been AI-generated or altered. The ability to circle questionable images on websites to see their provenance is coming to Chrome "later." Given how fast AI-generated imagery is flooding stock libraries and news feeds, this is overdue, though you can't help feeling that Google is very much trying to have its cake and eat it here: manufacturing many of the tools that create AI slop in the first place.

Search is also gaining "information agents" that can continuously monitor the web for updates on specific topics, pulling from blogs, news, and social media. These launch this summer for AI Pro and Ultra subscribers. 

Google Glass redux

Smart glasses have been Google's longest-running "coming soon" project, and at I/O 2026, they finally started coming into focus. Google showed an updated version of Project Aura, its smart glasses developed with XREAL, featuring a redesigned compute puck and new features including calendar and notes integration via Gemini.

Samsung's intelligent eyewear designs from Warby Parker and Gentle Monster also made their debut at I/O, both launching this fall. Like the base Ray-Ban Meta glasses, these will be audio-only with no display, supporting live translation, navigation via Gemini, and notification summaries. All of which makes whatever Apple might have up its sleeve in the field at next month's WWDC all the more interesting.

Vibe coding Android

Google will now let users build full native Android apps from text prompts in AI Studio, including an embedded Android emulator for previewing and editing, with the ability to publish directly to the Play Store. Apps can be exported to Android Studio, GitHub, or saved as a ZIP. Vibe-coding your way into the Play Store is a genuinely interesting development for smaller developers, though for consumers it could end up making the place more Wild West than ever.

Also interesting...

Gemini Spark is an always-on AI agent that runs continuously via virtual machines on Google Cloud, connecting to Google Workspace apps including Docs, Gmail, Sheets, and Slides, as well as third-party apps like Canva and Instacart. It's a version of OpenAI's operator-style agents, basically, albeit integrated into the Google workspace.

Google is also launching a Universal Cart, letting users add products from YouTube, Search, Gemini, and Gmail and check out across retailers including Nike, Target, and Walmart in one go.

Google's AI Ultra subscription, which launched at $250 per month last year, now starts at $100 per month, with a $200 per month tier giving access to Project Genie. That's a significant price cut, and one timed fairly transparently against OpenAI's equivalent tier.