AiGpu

Industry·

Figma Expands MCP Partnerships While Co‑Founder Flags Protocol Limits

Figma announced 20 new Model Context Protocol partners at its Config India event, yet co‑founder Dylan Field warns that MCP still has significant shortcomings, especially around asymmetric data sharing and security risks.

Figma logo alongside MCP protocol icons

At its first Config conference in India, Figma revealed a roster of 20 new Model Context Protocol (MCP) partners, including Amp, ClickUp, Linear, Notion, and Mistral. The move signals Figma’s commitment to opening its design data to AI agents that rely on MCP for tool‑calling and context exchange.

Despite the expansion, co‑founder Dylan Field reiterated his skepticism. In a media briefing he said, “MCP is not a perfect protocol” and highlighted that the standard currently favors frontier labs while leaving gaps in reciprocal context sharing. Field also pointed out that Anthropic, the protocol’s creator, does not expose its own cloud artifacts via MCP, creating an asymmetry that could limit broader adoption.

Why it matters for GPU and AI infrastructure

For GPU cloud providers like AiGpu, the evolution of open protocols such as MCP determines how smoothly AI workloads can tap into external data sources without custom integrations. A mature, balanced protocol reduces engineering overhead, improves workload portability, and lets GPU clusters focus on compute rather than plumbing.

Figma’s cautious stance—limiting data exposure to mitigate abuse—illustrates the tension between openness and security that infrastructure operators must navigate when offering multi‑tenant GPU environments.

  • aigpu
  • ai gpu
  • ai gpu cloud
  • aigpu dubai
  • mcp
  • figma
  • ai agents
  • protocol

By AiGpu Editorial · Editorial rewrite based on public reporting (TechCrunch AI)

← All articles