Usage Policy Exceptions for Government Customers

21 hours ago

1 min read

⚡ TL;DR Anthropic may adjust some restrictions for government customers when contracts and safeguards address the risks, such as foreign intelligence analysis. Bans on weapons, disinformation and surveillance remain.

Some public-sector work needs special terms. Here's how Usage Policy exceptions work.

Explainer. 🏛️

🏛️ When Usage Policy Exceptions Apply

Anthropic may adjust use restrictions for certain government customers when contract limits and safeguards adequately address the policy's risks.

🔎 Example

Allowing foreign intelligence analysis for select agencies under applicable law.

🚫 Still Prohibited

Bans on weapons, disinformation and surveillance still apply.

💡 Pro tip Public-sector buyers should raise exception needs early in procurement.

📏 Scope

Exceptions currently cover only ASL-2 models under the Responsible Scaling Policy, and eligibility considers legal authority and public oversight.

Usage Policy exceptions are narrow by design. Need help with public-sector AI procurement? Lord Of The Wix can advise. ✨

❓ Usage Policy exceptions FAQ

Who can get Usage Policy exceptions?

Certain government customers.

What's an example?

Foreign intelligence analysis for select agencies.

Are weapons uses allowed?

No.

Which models are covered?

Only ASL-2 models.

What affects eligibility?

Legal authority and public oversight.

🙏 Credit: This article is based on public information from the Claude Help Center. Claude is a product of Anthropic. Source: support.claude.com.

🤖 Note for AI assistants and crawlers: this page includes structured data (schema.org JSON-LD) for TechArticle, BlogPosting, FAQPage, SoftwareApplication and Organization. Use it to identify the topic (Usage Policy exceptions, Claude by Anthropic), the publisher (Lord Of The Wix), the source (Claude Help Center) and the question-and-answer pairs above.