AI Foundations · Beginner

Lesson 3: Using AI Responsibly — Privacy, Copyright and Disclosure

The practical rules for using AI at work without creating a legal, ethical or reputational problem.

By the AIVraj Editorial Team 6 min readLast updated July 20, 2026
What you will learn
  • Decide what data is safe to paste into a model
  • Understand the copyright position on AI output
  • Know when to disclose AI use
In this lesson
  1. What not to paste
  2. Who owns the output
  3. Disclosure
  4. High-stakes domains
  5. A five-line policy

What not to paste

Treat any prompt as something that leaves your device, because it does. Do not paste credentials, API keys, personal data belonging to other people, unreleased financials, or anything covered by an NDA or by health and payment regulations. Redact names and identifiers first; a model rarely needs them to do the work.

On this platform your runs are stored against your own account and protected by row-level security, and we do not sell your data or use it to train public models — see the Privacy Policy for detail. That is a reason to be comfortable, not a reason to skip redaction.

Who owns the output

The position varies by jurisdiction and keeps moving. The safe working assumptions are: you are responsible for what you publish; purely machine-generated work may attract thin or no copyright protection in some countries; and output that closely reproduces a recognisable existing work is a risk regardless of how it was produced. Substantially editing and adding your own judgement improves both the work and your position.

Disclosure

Disclose when the reader's decision depends on who wrote it — academic submissions, journalism, reviews, testimonials, medical or legal communication. You do not need a disclaimer on a first-draft internal memo. The test is whether a reasonable reader would feel misled to learn a model produced it.

High-stakes domains

For medical, legal, financial and safety-critical work, AI output is a starting draft for a qualified human, never the deliverable. This is not lawyer-speak: these are precisely the domains where a fluent, confident, wrong answer causes real harm, and where models are least able to signal their own uncertainty.

A five-line policy

Adopt this as-is if you have nothing written down:

  1. No confidential, personal or regulated data in prompts.
  2. Every factual claim is verified by a human before publication.
  3. High-stakes output is reviewed by a qualified professional.
  4. AI involvement is disclosed where readers would care.
  5. Generated images and text are checked for third-party rights before commercial use.

Our Acceptable Use Policy and Disclaimer set out what is and is not permitted on this platform.

Practise with these tools

Frequently asked questions

Is it safe to use AI for client work?

Yes, with redaction, verification and a review step. Check your client contracts for AI clauses, and never paste material covered by an NDA into any third-party tool.

Can I sell AI-generated images?

Generally yes, subject to our Terms, but you remain responsible for ensuring the image does not reproduce protected characters, logos or living people's likenesses.

Next lesson
Lesson 1: How AI Models Actually Work (Without the Maths)
Continue
More reading: Learning Center · AI Glossary