Google I/O 2025 event. It's an exciting time for AI development.

Google I/O 2025 took place from May 20-21, 2025.

The main in-person event was held at the Shoreline Amphitheatre in Mountain View, California.

The event was also broadcast online for a global audience.

Major New Capabilities & Integrations:

• Agentic AI Features (e.g., Agent Mode, Project Mariner): This is a big one. Gemini is becoming more proactive and capable of performing multi-step tasks for you across various apps and the web. Examples include:

  • Finding apartments, booking flights, or researching tickets by Browse, conducting research, and integrating with your Google apps.

  • "Buy for me" feature for shopping, allowing Gemini to monitor product prices and make purchases via Google Pay.

  • Deep Research: Enhanced capabilities to browse and analyze hundreds of websites in real-time, providing comprehensive research reports. You can also upload PDFs and images directly for analysis.

• Deeper Google App Integration: 

  • Gemini in Gmail, Calendar, Drive, etc.: You can now directly ask Gemini to search your emails, check your calendar, or find documents in your Drive, providing a more integrated experience for productivity.

  • Gemini in Chrome: A new feature to ask questions directly while Browse the web.

• Advanced Models and Reasoning: 

  • Gemini 2.5 Flash: Becoming the new default model, offering a balance of quality and speed.

  • Gemini 2.5 Pro (and Deep Think mode): Continues to be available for Gemini Advanced users, providing the most intelligent model for complex reasoning. Deep Think is an enhanced reasoning mode for even better responses.

• Creative & Multimedia Features: 

  • Veo & Imagen 4: New, state-of-the-art models for video and image generation are being integrated. Veo 3 is the first to have native support for sound effects, background noises, and dialogue. Imagen 4 offers improved image quality and text rendering.

  • Flow: An AI filmmaking tool to create cinematic scenes and stories.

  • AI Avatars in Vids: Generate polished video content with AI-driven avatars by just writing a script, eliminating the need for filming.

• Enhanced Collaboration & Learning: 

  • Canvas: A new interactive space within Gemini for creating, refining, and sharing documents and code in real-time.

  • Audio Overview: Transforms your documents into engaging, podcast-style discussions with AI hosts, summarizing material and drawing connections.

  • Quizzes: Easily create interactive quizzes based on documents or specific topics, with hints, explanations, and summaries.

• Accessibility Improvements: AI-generated descriptions for images (even without alt text) and the ability to ask follow-up questions about screen content for users who are blind or low-vision.

New Subscription Plans:

• Google AI Pro (formerly AI Premium): Priced at $20/month, offering higher rate limits and special features, including the Gemini app (Advanced), NotebookLM, and Flow.

• Google AI Ultra: A new, higher-tier plan (around $250/month, with potential initial discounts) that provides the most powerful models, highest rate limits, and early access to experimental AI features like Project Mariner and Veo 3.

It's clear that Google is pushing Gemini to be a comprehensive and proactive AI assistant integrated deeply across its ecosystem, moving beyond just conversational AI to actively help users with complex tasks.

References:

Official Google I/O Website:

• https://io.google/2025/explore/ - This is the main page for exploring the event content.

• https://io.google/ - The general Google I/O website.

Google's Official Announcements & Blog Posts:

• https://blog.google/technology/developers/google-io-2025-collection/ - A collection of news and announcements from Google I/O 2025 on the Google Blog.

• https://blog.google/technology/ai/io-2025-keynote/ - The Google I/O 2025 keynote announcement.

• https://developers.googleblog.com/en/google-io-2025-developer-keynote-recap/ - Recap of the Developer Keynote.