Google unveils ‘agentic era’ of AI with Gemini 3.5 launch

Google has unveiled what it describes as a major leap into the 'agentic era' of artificial intelligence, introducing a sweeping range of new AI-powered tools, products and infrastructure upgrades during its Google I/O 2026 developer conference in Mountain View, California.

Speaking during the event, Google CEO Sundar Pichai said the company is now shifting its focus toward bringing autonomous AI agents directly to everyday consumers after years of building the technology for developers and enterprises.

“Gemini 3.5 and Antigravity are unlocking a new world of agents and agentic capabilities. We’ve been bringing agents to developers and enterprises for a while. Now we are super focused on bringing the power of agents, safely and securely, to consumers so that it works for everyone,” said Pichai.

Google said the transformation builds on its long-term strategy of becoming an AI-first company, a journey it says began over a decade ago.

The tech giant noted that its approach combines custom-built hardware, research models, cloud infrastructure and consumer products into a single ecosystem designed to accelerate innovation at scale.

The company revealed that usage of its AI systems has exploded over the past two years. Google said the number of tokens processed by its AI models every month has surged from 9.7 trillion two years ago to more than 3.2 quadrillion today, reflecting growing adoption across products and developer platforms.

More than 8.5 million developers are now building experiences using Google’s AI models every month, while Google Cloud customers processed over a trillion tokens each during the past year.

Google also highlighted rapid growth across its consumer-facing products powered by Gemini integrations.

The company said AI Overviews in Search now reaches 2.5 billion monthly active users, while AI Mode surpassed one billion monthly users within a year.

Meanwhile, the Gemini app has crossed 900 million monthly active users, with users generating more than 50 billion images using built-in Nano Banana AI models.

Among the newly announced features is Ask YouTube, an AI-powered video discovery tool designed to help users navigate lengthy videos by surfacing the most relevant sections based on nuanced questions. Google said testing begins immediately ahead of a broader rollout in the United States later this year.

The company also introduced Docs Live, a voice-powered feature that allows users to verbally generate and edit documents without typing.

Google said the feature will roll out to subscribers this summer, with similar voice capabilities expected to arrive in Gmail and Keep shortly afterward.

To support the growing demand for AI services, Google announced massive infrastructure investments, saying annual capital expenditure is expected to rise to approximately $190 billion (approx. Ksh.24 trillion) this year, up from $31 billion (Ksh.4 trillion) in 2022.

Much of the spending will focus on custom silicon, including the launch of its eighth-generation Tensor Processing Units dubbed TPU 8t and TPU 8i.

Google also unveiled Gemini 3.5 Flash, a new AI model the company says combines frontier-level intelligence with significantly faster execution speeds.

According to Google, the model outperforms Gemini 3.1 Pro across several benchmarks while operating four times faster than competing frontier models.

The company added that the model could help large enterprises cut AI-related operational costs substantially.

The tech giant further announced Antigravity 2.0, a standalone desktop platform designed to manage cohorts of autonomous AI agents, as well as Gemini Spark, a personal AI agent capable of handling long-running background tasks and integrating with products such as Chrome, email and chat services.

Google said it is also introducing agentic capabilities directly into Search through Information Agents that can continuously gather insights and perform actions on behalf of users around the clock.

Additional tools previewed during the event included Daily Brief, Google Flow, Google Pics, Gemini-powered smart eyewear and Gemini for Science, an experimental AI research architecture aimed at accelerating discoveries in life sciences.

The announcements signal Google’s intensifying push to position Gemini at the centre of its next-generation AI ecosystem as competition in the artificial intelligence race continues to heat up globally.