$34M Series A<\/a> co-led by Accel and ICONIQ, bringing total funding to $38.5M. The core argument is structural: Figma stores designs in a proprietary WebGL format that agents must translate before anything ships, while Paper renders natively in HTML and CSS. ARR grew 25x in a single month after the Paper Desktop launch, with paying customers including Ramp, Vercel, and Y Combinator.<\/p>\nWhat this means for your business<\/h2>\n
The sharpest signal in this story isn’t the funding, it’s the MCP server comparison. Paper ships 24 bidirectional tools (MCP, or Model Context Protocol, is the standard that lets AI agents read from and write to external software) against Figma’s 3 read-only tools. If your engineering org is already routing AI coding agents like Claude Code through the design-to-development handoff, that gap decides whether agents are participants in your design workflow or just spectators of it. Teams still on Figma with heavy agent usage are the ones with an immediate exposure question.<\/p>\n
The “format is destiny” argument Paper is making deserves scrutiny. Figma is not standing still, and its own MCP investments are accelerating. But the structural disadvantage Paper is identifying is real: a proprietary canvas format was a moat in a world where humans translated designs into code manually. In an agentic workflow, that same format becomes friction. The 25x ARR growth in a single month is self-reported and almost certainly starts from a small base, but the directionality is credible given the benchmark data from Braintrust showing Paper agents running faster and consuming fewer tokens.<\/p>\n
The decision this actually reframes isn’t whether to switch design tools today. It’s whether your next design tooling contract, likely a Figma renewal, deserves the same default treatment it got three years ago. If your engineering culture is moving toward agent-first development, the renewal conversation should include a direct question about bidirectional agent access, not just feature parity. Figma can answer that question with product investment, but right now Paper holds the architectural advantage, and architectural advantages in platform software tend to compound before incumbents close them.<\/p>\n
Concept deep-dive: MCP bidirectionality<\/h2>\n
Model Context Protocol is the integration layer that lets an AI agent take actions inside a piece of software, not just read its contents. A read-only MCP connection means the agent can see your design canvas but can’t change it, similar to giving a contractor a floor plan they can study but not mark up. Bidirectional MCP means the agent can both read and write, making the design canvas a live workspace the agent participates in rather than a file it processes afterward. That difference determines whether AI agents accelerate your design cycle or just describe it.<\/p>\n
Based on reporting from Paper Raises $34M Series A From Accel and ICONIQ to Challenge Figma<\/a>, originally published 2026-07-30 20:28:00.<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"Share with your CTO Design startup Paper is betting that Figma’s dominance collapses once AI coding agents become the primary way software ships. The company closed a $34M Series A co-led by Accel and ICONIQ, bringing total funding to $38.5M. The core argument is structural: Figma stores designs in a proprietary WebGL format that agents […]<\/p>\n","protected":false},"author":1,"featured_media":7215,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[148],"tags":[207],"tmauthors":[],"class_list":["post-7214","post","type-post","status-publish","format-standard","has-post-thumbnail","category-ai-marketing","tag-cto"],"_links":{"self":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts\/7214","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/comments?post=7214"}],"version-history":[{"count":0,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts\/7214\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/media\/7215"}],"wp:attachment":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/media?parent=7214"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/categories?post=7214"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/tags?post=7214"},{"taxonomy":"tmauthors","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/tmauthors?post=7214"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}