Token Thrift
  1. Ask
  2. Agents and governance
  3. Security
  4. Your account
Started by Token Thrift

Security / 5 public sources

Can you use AI coding tools on regulated code like CUI or health data?

Short answer

Short answer: Only on specific plans with compliance settings turned on, and never with every feature. GitHub Copilot can be restricted to FedRAMP Moderate certified models on GitHub Enterprise Cloud with US data residency, but GitHub says Copilot itself is not yet FedRAMP authorized. Anthropic signs a BAA for its API and Enterprise plans once HIPAA settings are enabled, not for consumer plans, and connectors and several features stay outside the BAA.

Is GitHub Copilot FedRAMP authorized for CUI?

Not on its own yet. GitHub's docs say enterprises on GitHub Enterprise Cloud with US data residency can turn on a policy that limits Copilot to FedRAMP Moderate certified models, and its changelog says Copilot will become authorized through the data residency product's authorization path. The policy is off by default and changes pricing.

Does Anthropic sign a BAA for Claude?

Yes, for HIPAA ready services such as the first party API and Enterprise plans, after an owner enables HIPAA settings and accepts the BAA. Consumer plans are not covered.

Which Claude features are not covered by the BAA?

Anthropic lists connectors and MCP servers, several beta features, Cowork in the cloud, and the Claude Code VS Code and JetBrains extensions as not covered, even on a HIPAA ready Enterprise plan. Keep regulated data out of them.

What should you check before putting regulated data into an AI tool?

The exact plan and contract (BAA or FedRAMP scope), which settings must be on, which features are outside coverage, and data retention. Then get sign off from whoever owns compliance, in writing.

Written by Token Thrift from the public sources below. Vendor facts checked on November 2, 2026.

See also

Add your answer

Loading replies.

Add your answer

Don't paste anything your employer treats as confidential. Before anything is posted, we take out names of people, your company, its clients and internal projects, plus places and exact numbers, and show you the result to check. Your raw text is never stored. To redact it, we send it to Anthropic's API, which doesn't train on API data by default.

How Ask protects you

  • You post under a random handle. No name or employer is shown, and your email is never shown to anyone.
  • Before a reply is saved, a model removes names, employers, clients, places and exact figures, a fixed leak check reads it again, and you approve the result. Your raw text is never stored. It is sent to Anthropic's API for this step. Redaction lowers the risk; it is not a guarantee.
  • Follow-up questions are passed on by email, and neither person sees the other's address. You can answer privately.
  • A handle's first reply is reviewed before it is published.
  • Replies that are never published are deleted after 30 days. Deleting your handle removes your email; published replies stay up as anonymous text.

More in the privacy policy and the FAQ.

What you get for replying

  • A digest of everyone else's answers on that topic, at most one email a day, one click to stop.
  • A count of replies readers marked helpful, shown next to your handle.
  • Credit by handle in the monthly roundups when your answer is used.

Tips to stay hard to recognize

  • Describe your company by size and industry, not by name, product or city.
  • Leave out unusual tool combinations or internal project details only your team would know.
  • Round numbers and timelines yourself, even though redaction also does it.
  • Wait a while before writing about something that was just announced inside your company.
  • Redaction doesn't change your writing style, so a coworker with context could still recognize you.

Summary of public sourcesWritten by our research agent from 5 public sources. Not first-hand experience.

Teams that handle controlled unclassified information (CUI) or patient data ask which AI coding tools they may use at all. The short answer from vendor documentation is that it depends on the exact plan, settings and features, not on the product name. Compliance is opt in, and many features stay outside the covered scope even on the right plan.

For GitHub Copilot, GitHub's docs say an enterprise on GitHub Enterprise Cloud with US data residency can turn on a policy that limits Copilot to models with FedRAMP Moderate certification. The policy is off by default and changes Copilot pricing. GitHub's changelog says the model hosting is FedRAMP Moderate authorized and that Copilot itself will become FedRAMP authorized as part of the data residency product's authorization path, so Copilot is not yet authorized on its own. For Anthropic, its Privacy Center says a BAA covers HIPAA ready services such as the first party API and Enterprise plans, only after an owner turns on HIPAA settings and accepts it. Consumer plans are not covered. Even then, connectors, MCP servers, several beta features and the VS Code and JetBrains extensions are listed as not covered. Facts checked on GitHub's and Anthropic's pages on 2026-11-02.

Still open: whether contracting officers accept model level FedRAMP restrictions for CUI work, and how teams keep regulated data out of features that are available but not covered.

Sources

  • Vendor docs: FedRAMP-compliant models for GitHub Copilot | GitHub Docs
  • Vendor docs: Data residency (US + EU) and FedRAMP-authorized models now available in Copilot | GitHub Changelog
  • Vendor docs: Business Associate Agreements (BAA) for Commercial Customers | Anthropic Privacy Center
  • Reddit: Can GitHub Copilot be used for CUI and plugged into GitLab?
  • Reddit: What to check before a medical or billing office uses Claude