1 / 8At its annual developer conference in Mountain View, California, Google announced a range of new artificial intelligence-powered features across YouTube, Google Docs and developer platforms, as the company seeks to strengthen its presence in the fast-growing AI sector. Chief Executive Officer Sundar Pichai said the company is integrating AI into every part of its business, unveiling new coding tools, upgraded Gemini AI models and multimedia generation capabilities to compete with rivals such as OpenAI and Anthropic. (Image: AI generated)
2 / 8Gemini Omni (and Omni Flash): Google’s massive new foundational “world model” series designed to understand and simulate the physical world. It accepts multi-modal inputs (combining text, audio, images, and video) and outputs highly realistic video grounded in physics, allowing users to create content or conversationally edit their own videos by changing lighting, point-of-view, or elements while keeping characters consistent. (Image: AI generated)
3 / 8Gemini 3.5 Flash: The first in Google’s new 3.5 model series, explicitly optimized for agentic workflows, long-horizon tasks, and coding. It runs at four times the speed of comparable frontier models (while beating Gemini 3.1 Pro across major benchmarks) at a fraction of the cost, and is generally available starting today. (Image: AI generated)
4 / 8Google Antigravity 2.0: A major new “agent-first” development platform and standalone desktop app designed to orchestrate background AI subagents for parallel software engineering tasks. Driven by Gemini 3.5 Flash, it acts as a central hub where developers can delegate entire high-level coding goals, automatically spin up browsers to self-test and verify code, or use its new CLI and SDK to embed autonomous agent behaviors directly into their infrastructure. (Image: AI generated)
5 / 8Ask YouTube: A brand-new conversational search experience built directly into YouTube. Instead of old-fashioned keyword searching, users can type complex, conversational questions to get structured text responses alongside relevant videos that automatically cue up and jump straight to the exact timestamps answering the query. (Image: AI generated)
6 / 8Docs Live: A new voice-driven creation feature coming to Google Docs, Gmail, and Keep. It allows users to continuously speak their thoughts to let Gemini turn unstructured “brain dumps” into polished text, while also utilizing a full-screen voice wave interface in Keep to automatically separate rambling thoughts into distinct, organized notes. (Image: AI generated)
7 / 8Android Halo: A new, subtle visual status space coming to the top of Android device screens later this year. It integrates with Gemini Spark and other supported agents to provide at-a-glance visibility and progress live tracking of background AI tasks without forcing you to exit your current app. (Image: AI generated)
8 / 8Google AI Plan Restructuring : A major price drop and tier expansion for Google’s top-tier subscriptions. The flagship Google AI Ultra plan has been reduced from $250 to $200/month, and a new $100/month Ultra tier has been introduced, offering 5x higher Gemini app usage limits than the Pro plan alongside 20 TB of storage and YouTube Premium. (Image: AI generated)
