AI document collaboration is shifting from chat windows to superpowered docs
AI document collaboration is the emerging practice of blending shared documents, live data, and integrated AI assistants into a single workspace so teams can draft, analyze, and automate work without leaving the document surface, turning docs into active systems that coordinate tasks, decisions, and workflows for entire organizations. Coda’s decision to rebrand as Superhuman Docs is the clearest signal yet that the document, not the chatbot, is becoming the primary canvas for AI-powered work. Superhuman is rolling out a rebuilt Docs AI assistant, broader AI tool connections, and new data features for teams working across documents, workflows, and shared databases. The company even describes this as “Coda 5.0,” framing the move as an evolution rather than a rupture for existing users. In other words, this isn’t a cosmetic rename; it’s a bet that the future of AI collaboration lives inside documents, not in separate sidebars.
What makes Superhuman Docs more than a fancy text editor
Superhuman Docs is opinionated software: it assumes your team wants AI in the document, acting with full context, instead of hovering off to the side. Docs AI now sits inside the document surface itself, with access to team data, briefs, roadmaps, and connected tools, so it can behave more like a teammate than a generic bot. Users are already asking Docs AI to summarize sprints, create project trackers, draft decision records from meeting notes, compare pricing against competitors, and help new colleagues understand existing docs. The updated AI columns and AI blocks extend this intelligence across tables and pages, automatically summarizing content, surfacing action items, and finding themes in large datasets. According to Superhuman’s CEO Shishir Mehrotra, “AI is no longer a solo experience. Docs AI works with you the way a teammate would, with full context of everything your team has built.” For teams tired of copying text into chatbots, this is a tangible quality-of-life upgrade.
Claude and ChatGPT walk into a doc: MCP turns Superhuman into a hub
The most strategic Superhuman Docs feature is Docs MCP, which connects documents directly to external AI tools including Claude, ChatGPT, and Cursor. Through the Model Context Protocol, users can ask questions, request updates, and even build docs from their preferred AI tool while keeping the document as the shared source of truth. That matters: instead of forcing teams to pick one AI, Superhuman turns docs into neutral ground where multiple models collaborate around the same data. The company says Docs MCP was its fastest-growing feature during beta, with more than 40 improvements shipped in eight weeks. This is a quiet but bold stance in the AI-powered document tools market. Rather than chasing a monolithic “one agent to rule them all,” Superhuman positions docs as the orchestration layer for many agents. For organizations that already live in shared docs as operating systems, this is a more realistic upgrade than abandoning their workflow for a single, all-knowing chatbot.
ChatGPT Work: OpenAI’s agentic superapp takes aim at the same workloads
OpenAI’s launch of ChatGPT Work is a direct shot at the same enterprise document collaboration and productivity automation space. ChatGPT Work is a Codex-based agentic tool for knowledge workers, designed to move beyond answering questions and “get real work done across web, mobile, and desktop.” It kicks off complex workflows based on documents, spreadsheets, and other assets, and can pull data from third-party services like Slack, Microsoft Teams, Google Drive, SharePoint, calendars, and CRM systems. At the same time, Codex is merging into the broader ChatGPT desktop app, making Work and Codex the main modes while simple chat is relegated to the sidebar. OpenAI is clearly chasing a superapp strategy, folding the Codex and Atlas experiences into a single interface and beginning to sunset the standalone Atlas browser. Where Superhuman puts AI inside docs, OpenAI puts docs inside an agent that spans your entire device. ChatGPT Work vs Coda—now Superhuman Docs—is shaping up to be a contest between document-first and app-first visions of AI-powered work.
AI Views, Databases, and the next generation of document intelligence
Superhuman’s most ambitious bet is that documents will stop being static pages and become live, intelligent interfaces. AI Views, now in closed beta, let users describe the interface they need on top of live document data—a live seating chart built from a sales dinner registration list, a cost dashboard with sliders and heatmaps to model LLM spending, or a weekly standup tracker with leaderboard features. Changes made in an AI View flow through to other views in the document, so the interface is another way to work with the same underlying data. Superhuman Databases, also in closed beta, are designed to handle up to 1 million rows, with a single database able to connect to multiple docs. Together, these features push AI document collaboration into territory far beyond traditional editors or synchronous comments. While ChatGPT Work orchestrates tasks across tools, Superhuman Docs is turning the doc itself into a data-rich control panel for the organization. For enterprises, the real question isn’t which AI feels smarter—it’s which environment makes sustained, shared work feel less like fighting your tools and more like using them.
