Inside Google’s Beam Lab, an AI face appears
Not a prototype. Google just showed journalists a lifesize AI agent named Sophie that sees, speaks, and reads in real-time.
Why it matters
Google is moving embodied AI agents from research into physical deployments. This signals a major shift toward multimodal, vision-enabled conversational AI entering real-world environments—and raises immediate questions about how enterprises will adopt (and regulate) always-on AI agents.
The key facts
6 to knowGoogle Beam Lab: lifesize AI agent with vision, multimodal language, and body language
Agent named 'Sophie' demonstrates real-time vision (reads phones, papers, books)
Multilingual capability with attempted natural body language
Integration with Google services (maps, weather, restaurant recommendations)
Exclusive journalist access to Mountain View labs — first public demonstration
Published May 2026 — deployment-stage readiness signaling
Go to the source
The Verge Techtheverge.com
Publisher excerpt: They tell me no one outside Google has seen what we're about to see. No journalist has ever been in this building. Here in Google's Mountain View labs, the company's creating lifesize AI agents that can see you, and talk to you. This one's name is Sophie. It can speak any number of languages. It…