Google opened its annual developer conference Tuesday morning at the Shoreline Amphitheatre in Mountain View, California, turning the spotlight on its Gemini artificial intelligence lineup. The event, known as Google I/O 2026, drew developers and reporters eager to hear the company’s latest plans for AI search tools, autonomous agents, and new features in e-commerce.
The Verge’s on-site coverage captured the opening tone with a simple declaration: “It’s time to usher in Gemini season.” Reporters noted that the company faces growing competition from OpenAI’s ChatGPT and Anthropic’s Claude, and that Google must show clear differences in how its models perform on real-world tasks.
Attendees arrived expecting limited hardware announcements. Google had already unveiled the Googlebook laptop and Fitbit Air wearable the previous week, leaving the keynote to focus mainly on software and AI updates. Much of the Android news had also been shared earlier during a separate Android Show event.
According to the live blog, the company is preparing to discuss “vibe coding,” a term used internally to describe new AI-assisted development workflows. Observers said these tools could let programmers describe desired features in natural language and receive working code suggestions in return.
The keynote was scheduled to begin at 10 a.m. Pacific time, or 1 p.m. Eastern time. Organizers set up the usual large screens and demonstration areas inside the amphitheater, where Google has held the conference for many years.
Industry analysts watching from outside the venue said the pressure on Google has increased over the past two years as rival AI systems gained users quickly. One person familiar with the preparations told reporters that the company wants to prove Gemini can handle complex, multi-step tasks without constant human guidance.
Google executives are also expected to touch on smart glasses prototypes. While no firm release date has been confirmed, the live coverage indicated that concept demonstrations could appear during the keynote to show how the devices might integrate with Gemini models for real-time assistance.
The broader Silicon Valley conversation this week has centered on how AI agents might eventually handle shopping, travel planning, and customer service. Google’s presentation is likely to include early examples of these agents working inside its search and shopping products.
Developers in the audience said they are particularly interested in new application programming interfaces that could let third-party apps tap into Gemini’s reasoning capabilities. Several startups have already built experimental tools around earlier versions of the model and are waiting to see what updates arrive this week.
Officials at the event have not yet released detailed numbers on Gemini’s current user base or performance benchmarks. Those figures are expected to be shared during technical sessions scheduled for later in the week.
The conference continues through Friday with additional keynotes, workshops, and product demonstrations. Google has published a full schedule on its developer site for anyone unable to attend in person.
More announcements are anticipated in the coming days, particularly around updates to Google’s search experience and new commerce features that use AI to match buyers with products. The company has not confirmed whether any of these changes will roll out immediately after the event.