Three major AI companies -- normally rivals -- appear to have found some common ground. 

On Wednesday, Google announced via an X post that it will support Anthropic's Model Context Protocol (MCP). This protocol lets AI systems -- including agents -- access data stores, developer spaces, and business applications for better performance. Specifically, support will go to Gemini models and its software development kit (SDK). 

Also: Block's new open-source AI agent 'goose' lets you change direction mid-air

The news follows OpenAI's March 26 announcement that it would adopt MCP, first with its SDK. OpenAI CEO Sam Altman said MCP would also eventually be available for ChatGPT on desktop and in the app, though it is unclear when. 

Though the proliferation of AI agents -- assistants that help complete a range of tasks, sometimes autonomously -- seems endless (and increasingly impressive), they're only as good as the data they have access to. Especially in high-security and enterprise settings, agents need to be integrated with every system and data source individually, which can be tedious and hard to scale. 

MCP, which Anthropic open-sourced late last year, solves this by providing a single standard. Alongside the release, Anthropic provided pre-built servers for commonly used enterprise software, including Google Drive, GitHub, and Slack. 

Also: The plan to decentralize TikTok

"MCP is a good protocol, and it's rapidly becoming an open standard for the AI agentic era," said Demis Hassabis, co-founder and CEO of Google DeepMind. "We're excited to announce that we'll be supporting it for our Gemini models and SDK. Look forward to developing it further with the MCP team and others in the industry."

Other companies that have embraced MCP include Block, Apollo, Zed, Replit, Codeium, and Sourcegraph. The shift to using more open-source tools, including at the enterprise level, could indicate a larger sea change in the tech industry, especially as more companies invest in AI agents as the technology's next frontier. 

Hassabis did not clarify when Google would support MCP. 

Artificial Intelligence

The best AI for coding in 2025 (and what not to use - including DeepSeek R1)

I tested DeepSeek's R1 and V3 coding skills - and we're not all doomed (yet)

How to remove Copilot from your Microsoft 365 plan

How to install an LLM on MacOS (and why you should)

• The best AI for coding in 2025 (and what not to use - including DeepSeek R1)
• I tested DeepSeek's R1 and V3 coding skills - and we're not all doomed (yet)
• How to remove Copilot from your Microsoft 365 plan
• How to install an LLM on MacOS (and why you should)