25 of 42
ContentsPart IV · Chapter 12

Chapter 12Where do my words go?

Whose agreement is it?

It depends less on which model you use than on the arrangement around it: what kind of account, whose agreement, which features are on, and whose computer runs it.

The same model can sit inside very different arrangements. Four come up most.

Personal account

You sign up yourself, free or paid. The agreement is between you and the vendor. Vendors commonly give personal users a setting for whether their chats help train future models; Anthropic calls it "Model Improvement," and OpenAI's is "Improve the model for everyone."12 A paid personal plan is still a personal agreement.3

Organization account

Your employer or school signs the agreement with the vendor. Anthropic and OpenAI both say business data isn't used for training by default.456 Organization plans can add administrator controls such as single sign-on, audit logs, retention settings and role-based permissions, depending on the plan.7896

Regulated workspace

Some information, like patient records, is covered by specific laws. Using it needs more than an organization account: a specific contract and a deliberately configured workspace.1011

Your own hardware

A model you download and run yourself can work offline, so your words can stay on your machine.12 But there is no vendor to share the work: every safeguard, such as access control, audit records and protecting data in transit, is yours to provide.13

An introduction to the ideas, not legal advice and not the reference for any product's current terms. Vendors change their plans and features often; your organization's privacy and compliance office, your signed agreements and each vendor's current documentation decide what is allowed.

Go deeper

Terms change, so it's worth rereading them now and then. In 2024, staff at the U.S. Federal Trade Commission warned AI companies to honor their privacy commitments, and that quietly changing terms to use people's data in new ways could be unfair or deceptive.1415

Notes (15)
  1. 1Anthropic, “Is my data used for model training? (consumer),” Anthropic Privacy Center, March 16, 2026, https://privacy.claude.com/en/articles/10023580-is-my-data-used-for-model-training (accessed October 26, 2026).
  2. 2OpenAI, “Data controls in ChatGPT,” OpenAI Help Center, https://help.openai.com/en/articles/7730893-data-controls-in-chatgpt (accessed October 26, 2026).
  3. 3Anthropic, “Updates to our consumer terms,” August 28, 2025, https://www.anthropic.com/news/updates-to-our-consumer-terms (accessed October 26, 2026).
  4. 4Anthropic, “Is my data used for model training? (commercial),” Anthropic Privacy Center, August 18, 2026, https://privacy.claude.com/en/articles/7996868-is-my-data-used-for-training (accessed October 26, 2026).
  5. 5OpenAI, “Business data privacy, security, and compliance,” https://openai.com/business-data/ (accessed October 26, 2026).
  6. 6OpenAI, “Enterprise privacy at OpenAI,” January 8, 2026, https://openai.com/enterprise-privacy/ (accessed October 26, 2026).
  7. 7Anthropic, “Set up single sign-on (SSO),” Claude Help Center, https://support.claude.com/en/articles/13132885-set-up-single-sign-on-sso (accessed October 26, 2026).
  8. 8Anthropic, “Access audit logs,” Claude Help Center, June 15, 2026, https://support.claude.com/en/articles/9970975-access-audit-logs (accessed October 26, 2026).
  9. 9Anthropic, “Configure custom data retention controls for Enterprise plans,” Claude Help Center, https://support.claude.com/en/articles/10440198 (accessed October 26, 2026).
  10. 10U.S. Department of Health & Human Services, “Guidance on HIPAA & Cloud Computing,” HHS.gov, last reviewed December 23, 2022, https://www.hhs.gov/hipaa/for-professionals/special-topics/cloud-computing (accessed October 26, 2026).
  11. 11Anthropic, “HIPAA-ready Enterprise plans,” Claude Help Center, https://support.claude.com/en/articles/13296973-hipaa-ready-enterprise-plans (accessed October 26, 2026).
  12. 12LM Studio, “About LM Studio,” https://lmstudio.ai/docs/app (accessed October 26, 2026).
  13. 13U.S. Department of Health & Human Services, “Summary of the HIPAA Security Rule,” HHS.gov, last reviewed August 7, 2026, https://www.hhs.gov/hipaa/for-professionals/security/laws-regulations (accessed October 26, 2026).
  14. 14FTC Office of Technology, “AI Companies: Uphold Your Privacy and Confidentiality Commitments,” U.S. Federal Trade Commission (Technology Blog), January 9, 2024, https://www.ftc.gov/policy/advocacy-research/tech-at-ftc/2024/01/ai-companies-uphold-your-privacy-confidentiality-commitments (accessed October 26, 2026).
  15. 15FTC Office of Technology, “AI (and other) Companies: Quietly Changing Your Terms of Service Could Be Unfair or Deceptive,” U.S. Federal Trade Commission (Technology Blog), February 13, 2024, https://www.ftc.gov/policy/advocacy-research/tech-at-ftc/2024/02/ai-other-companies-quietly-changing-your-terms-service-could-be-unfair-or-deceptive (accessed October 26, 2026).

Chapter 12 · page 1 of 3

Turn pages with the ← and → keys.

Cite this page

Using this book

© 2026 Steve Sparks. All rights reserved. You're welcome to quote short passages and cite any page, with a link back. To reuse diagrams or larger parts, ask through GitHub.

Quoted material and figures from other sources remain their owners'.