Skip to main content

Acceptable Use Guidelines

The RCD LLM Service is a Clemson-hosted service for research and education workflows. Use of the service must remain consistent with Clemson University IT policies, Clemson AI guidance, and any sponsor, contract, or regulatory requirements that apply to your work.

In addition, use of the RCD LLM Service is limited to research and education. Non-academic personal and commercial use is prohibited.

Data Classification​

The data classification limits for the RCD LLM Service should match the current restrictions used for Palmetto and Indigo.

Please review the data classification guidelines to determine which category your data is in. Then refer to the table below.

CategoryAllowed on the RCD LLM Service?
Public✅ Yes
Internal Use✅ Yes
Confidential⚠️ Requires CCIT Security Approval
Restricted⚠️ Requires CCIT Security Approval
tip

If you need help classifying your data, please contact CCIT Security through email support.

Research And Sponsored Work​

Unpublished research data is not automatically prohibited on this service. However, you remain responsible for making sure the data is permitted under the classification table above and under any additional restrictions that apply to your project.

In particular, check whether your work is subject to:

  • grant or contract restrictions
  • data use agreements
  • IRB or human subjects restrictions
  • export control or CUI requirements
  • collaborator or publisher confidentiality obligations

If those requirements are stricter than the table above, follow the stricter requirement.

General Use Expectations​

When using the RCD LLM Service:

  1. Do not enter data above the approved classification level for your project.
  2. Do not paste secrets such as passwords, private keys, API tokens, or other credentials into prompts.
  3. Review all model outputs carefully before relying on them in research, software, reports, or other deliverables.
  4. Use the service in ways consistent with Clemson University AI Guidelines and IT Policies and Standards.

Output Quality And Responsibility​

Like other LLM services, this system can produce inaccurate, incomplete, or misleading output. Users are responsible for:

  • checking factual claims, citations, and calculations
  • reviewing generated code before running it in production or on shared systems
  • making appropriate authorship, disclosure, or attribution decisions for research, coursework, and publications