Google launches Gemini 3.5 Flash, focusing on AI agents over chatbots
Original: With Gemini 3.5 Flash, Google bets its next AI wave on agents, not chatbots
Why This Matters
Signals major shift in AI development from chatbots to autonomous agent systems
Google introduced Gemini 3.5 Flash at I/O 2026, a new AI model designed for coding and autonomous agents. The model can execute coding pipelines independently, manage research projects, and build operating systems from scratch with minimal human input.
Google unveiled Gemini 3.5 Flash at its annual I/O developer conference, marking a strategic shift from conversational AI to agentic AI tools. The model outperforms Google's 3.1 Pro on coding, agentic tasks, and multimodal reasoning benchmarks while being four times faster than other frontier models. An optimized version runs 12 times faster with the same quality. DeepMind's Koray Kavukcuoglu highlighted the model's ability to run multiple AI agents simultaneously on long-running tasks. Google demonstrated agents building a complete operating system within Antigravity, the company's agentic development platform. The model can run autonomously for hours, pausing only for human input at decision points. Google also released Antigravity 2.0 as a standalone desktop application for agent-first development. Partners including banks and fintechs are already using the model to automate multi-week workflows.