Google opens Google Home to AI agents via MCP

// 2h agoNEWS

Google opens Google Home to AI agents via MCP

Google is opening its Google Home ecosystem to outside AI agents via the Model Context Protocol (MCP). This integration allows third-party AI models like Claude or custom agents to monitor smart home devices and take actions, effectively turning the smart home into a new integration surface for AI capabilities.

// ANALYSIS

This marks a significant shift in smart home ecosystems, moving away from closed, first-party AI assistants toward open integration with diverse AI models. It introduces the Model Context Protocol to the consumer smart home space and raises critical considerations for security, permissions, and the potential "blast radius" of autonomous AI actions in the physical world. This could significantly accelerate the utility of smart home devices by enabling more complex, cross-platform AI-driven automation.

// TAGS

googlegoogle-homemcpagentsmart-home

DISCOVERED

2h ago

2026-09-17

PUBLISHED

2h ago

2026-09-17

RELEVANCE

8/ 10

AUTHOR

SreeramG

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

INFRA1h ago

B.AI stacks 90% subsidy on DeepSeek multimodal

AI compute distribution platform B.AI has rolled out a promotional "double subsidy" offer following the release of DeepSeek's native multimodal Mixture-of-Experts (MoE) model. By stacking a 90% platform-level discount on top of DeepSeek's official price reductions, B.AI allows developers and AI agents to access frontier multimodal inference at 10% of standard pricing.

MODEL2h ago

Google launches Gemini 3.8 Live with extended thinking

Google has launched Gemini 3.8 Live with real-time speech, vision, and extended thinking across 97 languages. The model can run tools and process complex tasks asynchronously in the background while maintaining an unbroken voice conversation with the user.

MODEL4h ago

Manycore Tech open-sources UniPhysGen for robotic simulation

Manycore Tech, in partnership with Zhejiang University and UESTC, has open-sourced UniPhysGen, a physical AI model that converts raw 3D assets into interactive, simulation-ready digital twins by predicting articulation kinematics and physical properties. Alongside model weights and code, the team released the UniPhys-40K dataset and UniPhys-Bench evaluation benchmark to advance embodied AI research.

Google opens Google Home to AI agents via MCP