Is there finally a Google search engine with insider knowledge for enterprises… but with OpenAI? It certainly seems that way.
Today OpenAI does launched company knowledge in ChatGPTan essential recent feature for subscribers to paid ChatGPT Business, Enterprise and Edu plans that allows them to call up company data directly from third-party work apps including Slack, SharePoint, Google Drive, Gmail, GitHub, HubSpot and combine it in ChatGPT results.
As OpenAI Apps CEO Fidji Simo wrote in a post on the X social network: “It brings together all the context from your apps (Slack, Google Drive, GitHub, etc.) into ChatGPT so you can get answers specific to your business.”
Interestingly, OpenAI’s blog post about the feature states that it is “based on a version of GPT-5 trained to look at multiple sources to provide more comprehensive and accurate answers,” which sounds to me like a recent, improved version of the family of models the company released in August, although there are no additional details on how it was trained.
However, enterprise knowledge in ChatGPT is distributed globally and aims to make ChatGPT a central point of access to verified organizational information, supported by secure integrations and enterprise-level compliance controls, and to provide employees with much faster access to company information while working.
Now, instead of switching to Slack to find a task and instructions you received, or going to Google Drive and opening specific files to find the names and numbers you want to call, ChatGPT can deliver all this type of information directly to your chat session – if your company allows the appropriate connections.
As OpenAI COO Brad Lightcap – wrote X in a post on the social networking site: “The company’s expertise has changed the way I use chatgpt at work, more than anything we’ve built before – let us know what you think!”
It builds on third-party app connectors unveiled in August 2025, although these were only intended for individual users on ChatGPT Plus plans.
Connecting ChatGPT to workplace systems
Enterprise teams often face the challenge of data fragmentation across various internal tools – email, chat, file storage, project management and customer platforms.
The company’s expertise bridges these silos, enabling ChatGPT to connect to approved systems, such as and other supported applications, through enterprise-managed connectors.
Each response generated from the company’s knowledge includes citations and direct links to the original sources, allowing teams to see where specific details came from. This transparency helps organizations maintain data reliability while increasing productivity.
OpenAI confirms that the company’s expertise leverages a version of GPT-5 optimized for cross-source inference and cross-system synthesis, ensuring detailed, contextually right results even from disparate sources.
Built with enterprise control and security in mind
The company’s expertise has been designed from the ground up with corporate governance and compliance in mind. Respects existing permissions in connected applications – ChatGPT can only access what the user already has permissions to, and by default never exercises on corporate data.
Security features include industry-standard encryption, support for single sign-on and SCIM for account sharing, and whitelisting of allowed IP addresses to restrict access to approved corporate networks.
Enterprise administrators can also define role-based access control (RBAC) policies and manage permissions at the group or department level.
The OpenAI Enterprise Compliance API provides a full audit trail, allowing administrators to view call logs for reporting and regulatory purposes.
This feature helps enterprises meet internal governance standards and industry-specific requirements such as SOC 2 and ISO 27001 compliance.
Administrative configuration and connector management
For enterprise deployment, administrators must enable enterprise knowledge and its connectors in the ChatGPT workspace. Once connectors are activated, users can authenticate their own accounts for each work app they need to access.
On Enterprise and Edu plans, connectors are disabled by default and require explicit admin approval before employees can exploit them. Administrators can selectively enable connectors, manage access by roles, and require single sign-on authentication for increased control.
Business plan users, on the other hand, have connectors automatically enabled if they are available in their workspace. Administrators can still oversee which connectors are approved, ensuring compliance with internal IT and data policies.
Company knowledge becomes available to any user with at least one energetic connector, and admins can configure group-level permissions for different teams—for example, limiting engineers’ access to GitHub while enabling Google Drive or HubSpot for marketing and sales purposes.
How the company’s knowledge works in practice
Activating your company’s knowledge is plain. Users can start a recent or existing conversation in ChatGPT and select “Company Knowledge” under the message composer or from the tools menu.
Once connected apps are authenticated, they can ask questions as usual, such as “Summarize the latest views and risks associated with this account” or “Compile a summary of Q4 results from project trackers.”
ChatGPT searches connected tools, retrieves the appropriate context, and generates a response with full citations and source links.
The system can combine data from different applications—for example, combining Slack updates, Google Docs notes, and HubSpot CRM records—to create an integrated view of a project, client, or initiative.
If company knowledge is not selected, ChatGPT may still exploit connectors to a circumscribed extent within the default experience, but responses will not contain detailed citations or synthesis from multiple sources.
Advanced exploit cases for enterprise teams
For development and operations leaders, company knowledge can act as a centralized layer of intelligence that displays real-time updates and dependencies in convoluted workflows. For example, ChatGPT can summarize open GitHub pull requests, highlight unresolved line issues, and link to engineering discussions in Slack – all in one result.
Technical teams can also exploit it for incident retrospectives or release planning, pulling relevant information from issue trackers, logs and meeting notes. Procurement and finance leaders can exploit it to consolidate purchase requests or budget updates on shared drives and internal communications.
Because the model can simultaneously reference structured and unstructured data, it supports a variety of scenarios, from compliance documentation reviews to cross-department performance summaries.
Privacy, data storage and compliance
Enterprise data protection is a central element of enterprise knowledge. ChatGPT processes data according to OpenAI’s enterprise-grade security model, ensuring that no connected application data leaves the secure perimeter of an organization’s authorized environment.
Data retention policies vary by connector. Some integrations, such as Slack, support region-specific data storage, while others – such as Google Drive and SharePoint – are available to US customers, with or without data storage. Organizations with regional compliance obligations can obtain detailed information on connector-specific security documentation.
No geographical restrictions apply to the company’s knowledge, making it suitable for international organizations operating in multiple jurisdictions.
Limitations and future improvements
Currently, users must manually enable company knowledge in each recent ChatGPT conversation.
OpenAI is developing a unified interface that automatically integrates the company’s knowledge with other ChatGPT tools – such as viewing and generating graphs – so users won’t have to switch between modes.
Once enabled, Company Knowledge temporarily disables web browsing and the generation of visual results, although users can switch modes within the same conversation to re-enable these features.
OpenAI also continues to expand the network of supported tools. Recent updates have added connectors for Asana, GitLab Issues, and ClickUp, and OpenAI plans to support future MCP (Model Context Protocol) connectors to enable custom integrations from developers.
Some essential details regarding knowledge about the company remain unclear based on OpenAI’s published materials. There is no word yet on whether the system will be able to detect and exclude information marked as sensitive, whether organizations will be able to separately enable or disable data training for this feature, or whether users will ultimately be able to choose which model supports this feature.
OpenAI also did not say whether this version of GPT-5 is recent or specific to this feature, or what service level guarantees exist to ensure accuracy and prevent hallucinations in company-specific responses. VentureBeat has emailed OpenAI advocates with these and related questions and is awaiting a response, which we will post if and when we receive one.
Availability and start-up
Corporate knowledge is now available to all ChatGPT Business, Enterprise and Edu users. Organizations can start by enabling this feature in the ChatGPT messaging tool and connecting approved apps to work.
For enterprise deployments, OpenAI recommends a phased rollout: first enabling basic connectors (such as Google Drive and Slack), configuring RBAC and SSO, and then expanding to specialized systems once data access policies have been verified.
Procurement and security leaders evaluating this feature should remember that the company’s knowledge is covered by existing ChatGPT Enterprise terms and benefits from the same encryption, compliance and service level guarantees.
With the company’s expertise, OpenAI aims to make ChatGPT not just a conversational assistant, but an wise interface to enterprise data – delivering secure, contextual insights that aid technical and business leaders operate with confidence.