For teams

ChatGPT Enterprise privacy: what it protects — and what it doesn't

Last updated: November 17, 2026

The short answer

ChatGPT Enterprise is OpenAI's plan for larger companies. It's much more private than the personal plans: OpenAI says it doesn't train on your business data by default, and you get company login, encryption and control over how long chats are kept. What it doesn't do is keep your data from reaching OpenAI — everything your staff type still goes to OpenAI's servers.

What ChatGPT Enterprise gives you

According to OpenAI's enterprise privacy page:

  • No training on your business data by default.
  • You own your inputs and outputs (where the law allows).
  • You control how long data is kept.
  • Company login through SAML single sign-on, and control over who can use which features.
  • Encryption at rest (AES-256) and in transit (TLS 1.2+), and a completed SOC 2 audit.

OpenAI doesn't publish an Enterprise price; you get a quote from its sales team. Smaller teams can use ChatGPT Business, which has the same no-training default.

What it doesn't cover

  • Your data still leaves the building. Client names, contracts and patient details are sent to and stored by OpenAI, even if they aren't used for training.
  • Admins can see chats. Workspace admins and compliance tools may be able to access and export what staff typed — fine for oversight, awkward for anything confidential between colleagues and clients.
  • Only ChatGPT is covered. Staff who also use Claude, Gemini, Copilot or a personal ChatGPT account are outside it.
  • Mistakes still happen. Someone can still paste a password or a full customer list. Enterprise records it; it doesn't stop it.

Do you need Enterprise, Omna, or both?

  • Enterprise is about OpenAI's promises: no training, company login, retention rules.
  • Omna is about what's sent: private details never leave the computer, in any AI tool, with a report you can show clients.
  • Many teams use both. If you're not ready for an Enterprise contract, Omna plus a written policy (try our free template) covers the biggest risk today. See also AI data loss prevention.

Sources

Note: based on the companies' own pages as of November 17, 2026. AI companies change their plans and policies often — check their current pages before relying on this. This is general information, not legal advice.