CopilotKit raised $27 million in Series A funding led by Glilot Capital, NFX, and SignalFire to commercialize AG-UI, the open-source protocol that lets AI agents render interactive interfaces inside enterprise apps rather than dumping walls of text into a chat window. The Seattle-based startup, founded by brothers Atai and Uli Barkai, says AG-UI and CopilotKit together see millions of installs per week, with a large share of the Fortune 500 already running the stack in production. The round funds a push into self-hosted enterprise deployments, the part of the agent market most cloud-native rivals have left underserved.
AG-UI standardizes how agents talk to user interfaces — streaming chat, front-end tool calls, and shared state for human-in-the-loop workflows. It sits alongside the Model Context Protocol (MCP) and Agent2Agent (A2A) protocol in the emerging agent stack, and is supported by Google, Microsoft, Amazon, and Oracle on the infrastructure side, plus framework vendors LangChain, Mastra, PydanticAI, and Agno. CopilotKit's enterprise customer list already includes Deutsche Telekom, Docusign, Cisco, and S&P Global.
The pitch to developers is generative UI. Instead of an agent answering a revenue question with a paragraph, it can render a pie chart built from the company's own design system that the user can click into. "The agent can reply to you, not just with blocks of text, but with interactive UIs that are defined by your own company," CEO Atai Barkai said.
Key facts
- 01CopilotKit raised $27M in Series A funding led by Glilot Capital, NFX, and SignalFire.
- 02AG-UI and CopilotKit see millions of installs per week, with Fortune 500 deployments in production.
- 03Enterprise customers include Deutsche Telekom, Docusign, Cisco, and S&P Global.
- 04AG-UI is supported by Google, Microsoft, Amazon, Oracle, LangChain, Mastra, PydanticAI, and Agno.
- 05The Seattle-based startup has 25 employees and plans to use the round to grow the team.
He framed the pain point in plain terms. "If, for example, a user asks for breakdown of revenue by category, instead of getting this kind of big, impenetrable paragraph, you get a pie chart, and it's your own design of the pie chart that the user can interact with." Developers control how much latitude the agent has — pixel-perfect lockdown on one end, freeform composition from a component catalog on the other.
“AG-UI and CopilotKit see millions of installs per week, with the open-source layer designed so 95% of users build for free while the top enterprises pay.”— Jaeden Schafer
Alongside the funding, CopilotKit is launching CopilotKit Enterprise Intelligence, a self-hostable bundle aimed squarely at companies that won't push agent traffic through a third-party cloud. That positioning is the company's main wedge against Vercel's AI SDK, the assistant-ui component library, and OpenAI's Apps SDK — the last of which only renders rich interfaces inside ChatGPT itself.
Atai Barkai argued the difference is horizontal versus vertical integration. "If there are two things we hear in almost every single enterprise conversation, enterprises want optionality and they want self-hosting," he said. "Maybe they're already using the Google, Amazon, Oracle, Microsoft, LangChain, Mastra stacks. They want optionality, and they want self-hosting, and these are two things that they don't really get in the Vercel stack."
The open-source-plus-enterprise model carries a familiar tension: the protocol has to stay a neutral standard while the company sells software on top of it. Uli Barkai, head of growth, said the two sides are designed to reinforce each other. "They're very much complementary. Our strategy is to be the default choice in the ecosystem, and then to monetize the top enterprises."
He went further on the free tier. "So it's very much in our interest that the open source is the best out there, and the 95% of users can just go build and get started without paying anyone or talking to anyone." That mirrors the playbook that has worked for Vercel, HashiCorp, and Databricks — but it has also broken down for plenty of others when the commercial layer started cannibalizing the community one.
Skeptics will note that the agent UI category is crowded and early. Vercel has distribution, OpenAI has the Apps SDK locked to ChatGPT, and the major clouds backing AG-UI today could just as easily ship competing primitives tomorrow if the protocol gains enough leverage to matter. CopilotKit's 25-person team is small relative to the surface area it has to defend across frameworks, clouds, and enterprise procurement cycles.
Still, the funding lands at a moment when enterprises are visibly tired of chatbot-shaped AI. The interesting bet inside the $27M round is not that agents will reach more apps — that's already happening — but that the interface layer between agents and humans will standardize on something open rather than something owned by a single cloud. If AG-UI holds that position, CopilotKit becomes the default commercial on-ramp for a category every Fortune 500 will eventually have to buy. If it doesn't, the company will be a thin layer on top of someone else's stack.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
