🐥Highlights from Google I/O

Hey clever people, 

Here’s your quick roundup of AI-related announcements unveiled at Google I/O:

1. Agent Mode in the Gemini App

You can now delegate complex planning to Gemini. Agent Mode combines live web access, deep research, and Google app integrations to handle multi-step tasks with minimal input. (Be the HR in your workspace)

2. Google Veo 3

The latest AI video model generates high-quality videos with audio. It’s better at understanding physics, syncing lip movements, and responding to both text and image prompts. (No more retakes, no more production houses, all you need to get your next viral reel is Google VEO 3) 

3. Jules Code Assistant (Beta)

Meet your new coding sidekick. Jules understands project context, links to your codebase, and can autonomously write tests, build features, and fix bugs. (When life gives you bugs, you need Jules' code assistant)

4. AI Filmmaking Tool: Flow

Designed for use with Veo, Imagen, and Gemini, Flow lets creators generate cinematic scenes from prompts and manage their projects in one place. (No need for a drone when you can prompt the view.)

5. Project Astra (Live)

Now built into Search and the Gemini app, Astra can use your phone’s camera to have real-time, back-and-forth conversations about what it “sees.” (Your inbuilt guide for your day-to-day life) 

6. Imagen 4

Google’s new image model brings sharper detail, supports photorealistic and abstract styles, handles up to 2K resolution, and even improves spelling and typography in images. (Graphic designers' AI best friend is here) 

7. Deep Think Mode in Gemini 2.5 Pro

A major upgrade for reasoning tasks. This mode tackles complex coding and math using parallel thinking. 

8. Gemini 2.5 New Capabilities

Beyond Deep Think, Gemini 2.5 now supports native audio generation and computer control, expanding what it can do across devices.

From: @dr_cintas

Latest news dropping in your inbox soon,

 — Written by Aaron & The Clever Nest Team