OpenAI launches Company Knowledge for ChatGPT Business, Enterprise, and Edu—securely linking Slack, Drive, SharePoint, and GitHub for context-aware AI answers.

In the fast-paced world of modern business, where information is scattered across multiple digital tools, finding the right mix can feel like piecing together a puzzle. OpenAI’s recent unveiling of Company Knowledge addresses this very challenge. This feature equips ChatGPT Business, Enterprise, and Edu users with the ability to draw directly from internal company data, delivering answers that are not only precise but also deeply rooted in an organization’s unique ecosystem. For teams navigating complex projects or urgent decisions, it’s a step toward reclaiming time lost to endless tab-switching and fragmented searches—turning AI from a helpful sidekick into a reliable colleague who knows the full story.

This development builds on OpenAI’s ongoing efforts to embed ChatGPT more seamlessly into professional routines. Earlier in August 2025, the company introduced third-party app connectors for individual users on Plus plans, laying the groundwork for broader integrations. Now, with Company Knowledge, the focus shifts to enterprise-scale utility, where scattered documents and conversations become a unified resource.

From Standalone Assistant to Connected Knowledge Hub

At its core, Company Knowledge redefines how ChatGPT interacts with workplace information. Users can now enable the feature at the start of a conversation, prompting the AI to query connected applications in real time. This isn’t a simple search; it’s a synthesis process where ChatGPT pulls threads from multiple sources to weave coherent responses.

The feature’s design emphasizes accessibility without disruption. Once activated via a toggle under the message composer, ChatGPT scans apps like Slack for threaded discussions, Google Drive for collaborative files, SharePoint for shared repositories, and GitHub for code-related insights. Outputs include clickable citations that link straight to the source material, allowing users to verify details instantly. This transparency fosters trust, especially in environments where accuracy can make or break a deal. OpenAI notes that responses are generated with date filters and the ability to “think while it searches,” ensuring time-sensitive information, such as recent project blockers, rises to the top.

Consider a product manager querying outstanding tasks for a feature release: ChatGPT might review GitHub repositories for open TODOs, cross-reference Linear tickets, and scan Slack engineering channels for bug reports, then summarize what’s tracked, what’s pending, and what needs logging. Such examples highlight how this capability streamlines workflows, reducing the cognitive load on knowledge workers who often dedicate up to 19% of their week to information hunting, according to industry studies on enterprise productivity.

GPT-5’s Role in Multi-Source Reasoning

Powering these interactions is a specialized version of GPT-5, fine-tuned specifically for cross-platform analysis and response generation. This isn’t a public rollout of the full GPT-5 model but an enterprise-optimized variant trained to handle retrieval from diverse data streams while prioritizing factual grounding. It excels at multi-source synthesis—combining unstructured text from emails with structured data from project trackers—to deliver balanced overviews rather than isolated facts.

OpenAI describes this as enabling “more comprehensive and accurate answers,” where the AI doesn’t just regurgitate content but contextualizes it. For ambiguous queries, like “What are the key risks in our Q4 launch plan?”, it scans discussions across tools, highlighting consensus points alongside dissenting views or unresolved items. This approach mirrors how human teams deliberate, providing not a polished verdict but a reflective summary that invites further input. As Fidji Simo, CEO of Instacart and an early design partner, posted on X, “it brings all the context from your apps… together in ChatGPT so you can get answers that are specific to your business.”

A Growing Web of Integrations

Company Knowledge launches with a robust set of connectors, reflecting OpenAI’s commitment to interoperability across popular enterprise stacks. Supported applications include Google Drive (encompassing Docs, Sheets, and Slides), Microsoft SharePoint and Teams, Slack, Asana, GitHub, ClickUp, Dropbox, Box, Gmail, Outlook, HubSpot, and Salesforce. These integrations allow ChatGPT to operate as a neutral aggregator, indifferent to whether a team favors Google Workspace or Microsoft 365.

For organizations with bespoke needs, OpenAI provides pathways for custom setups through the Model Context Protocol (MCP) and developer tools, enabling secure exposure of internal databases or APIs. Administrators handle initial connections, ensuring seamless onboarding without individual user friction. This modular setup positions ChatGPT as an overlay layer—one that indexes and reasons over existing systems rather than replacing them—making it adaptable for hybrid environments common in global firms.

Safeguarding Trust: Security and Governance

In enterprise settings, innovation without ironclad security is a non-starter. OpenAI has woven compliance into Company Knowledge’s fabric from day one. The feature adheres strictly to users’ existing permissions: ChatGPT accesses only what an individual is authorized to see within each app, preventing any overreach. Data remains within the organization’s secure boundaries, processed with end-to-end encryption both in transit and at rest.

Administrative controls are equally robust. Enterprise and Edu admins can toggle connectors, set role-based access policies at group or department levels, and monitor usage via the Compliance API, which logs interactions for auditing. Additional layers include Single Sign-On (SSO) for authentication, SCIM for user provisioning, and IP allowlisting to confine access to corporate networks. OpenAI underscores that no customer data ever trains its models, a policy that aligns with data sovereignty requirements under regulations like GDPR, CCPA, and HIPAA.

These measures stem from feedback by early partners, who prioritized least-privilege access and traceable sources. The result is a system that empowers without exposing, allowing teams to experiment confidently while maintaining oversight.

Navigating the Competitive Landscape

OpenAI’s timing couldn’t be sharper, arriving amid a surge in tools vying for enterprise mindshare. Microsoft’s Copilot draws from the Microsoft Graph across Outlook, SharePoint, and Teams, while Google’s Gemini for Workspace integrates with Docs and Gmail for similar contextual pulls. What sets Company Knowledge apart is its vendor-agnostic stance: it spans ecosystems, suiting organizations with mixed-tool arsenals—GitHub for developers, Asana for marketers, SharePoint for operations.

Practical benchmarks from OpenAI’s trials show efficiency gains in real-world scenarios. A legal team might query past NDA clauses, pulling from archived Drive files; a finance analyst could assemble Q3 expense trends from spreadsheets and dashboards, all with verifiable links. For ambiguous asks, like surfacing 2026 R&D priorities, it reviews memos and threads to outline agreements and gaps, aiding nuanced discussions.

Currently, activation is manual per conversation, and it pauses web browsing or image generation to focus on internal sources—a deliberate choice for depth over breadth. Users can toggle it off mid-chat to access those functions without losing context.

Looking Ahead: Toward a Seamless AI Workspace

OpenAI plans to automate activation and fuse Company Knowledge with ChatGPT’s broader toolkit, including browsing, code interpretation, and visualization, in the months ahead. Imagine analyzing a Sheets model, generating a chart, and drafting a deck—all citing internal files in one flow. This evolution hints at an “AI intranet,” where ChatGPT becomes the live pulse of corporate memory.

For leaders, the takeaway is clear: in an era where data is both asset and liability, tools like this don’t just accelerate tasks—they humanize them, surfacing insights that feel intuitive and owned. As organizations lean into this shift, the focus turns to thoughtful integration: selecting connectors that fit, training teams on citation verification, and using audits to refine access. Company Knowledge isn’t about replacing human judgment; it’s about amplifying it, ensuring every decision draws from the collective wisdom already at hand.

In a workplace where collaboration spans borders and tools, this feature arrives as a quiet revolution—one that makes the invisible labor of context-building visible and effortless. As Lightcap put it, it’s changed how he works more than anything else OpenAI has built. For businesses ready to connect the dots, it’s an invitation to work not harder, but with the full picture in view.