AI Disclosure
Last updated November 12, 2026
Which parts of AI Mastery Academy are produced by AI, where it can be wrong, what is sent to the model providers, and what it never decides. The course material you pay for is written and edited by people.
1. What is written by people
The course itself is human-written. Lessons, worksheets, projects and rubrics are authored and edited by the academy before they are published, and a lesson only becomes visible to you once it has passed the academy's own quality checks.
2. What uses AI at the moment you use it
- The study assistant, which answers questions about the lesson you are reading.
- Generated practice: quizzes, flashcards, study plans and exam questions built from the lesson material.
- Feedback on project submissions, scored against the rubric published on the project page.
- Interview answer review and code explanation in the practice tools.
- Audio narration of lesson text, if you choose to listen instead of read.
3. AI output can be wrong
Generated answers, quizzes and feedback are produced automatically and can be incomplete, out of date or simply incorrect. Treat them as a study aid, not as an authority. Check anything factual against the lesson text or a source you trust before you rely on it, especially for legal, medical, financial or safety-related questions. The lesson pages are the reference material; the assistant is there to help you read them.
4. What is sent to the AI providers
To produce a result, the text relevant to your request is sent to a third-party AI provider: the lesson passage you asked about, your prompt, your notes or your project description. Those providers return the result and are not permitted to use your content to train their own models. Please do not paste confidential business information, or personal information about other people, into any AI feature on this platform.
5. How graded work is assessed
Project and capstone submissions get a score from an automated review that works through the same rubric a person would use, and the per-criterion breakdown is shown to you. When a result falls near the pass mark, it is reviewed a second time independently; if the two reviews disagree, the submission is flagged for a human to confirm before it counts. A score you believe is wrong can be raised with support, and a person will look at it.
6. What AI never decides
- Whether you have paid, and whether your access is open: that is decided by your membership record and the payment provider's confirmation, never by a model.
- Whether your account exists, what role it has, or whether it is removed.
- Whether a certificate is issued: eligibility is checked against your recorded progress.
7. Reporting a bad answer
If an AI feature gives you something wrong, unhelpful or inappropriate, email support@aiacademy101.com with the lesson or page you were on. It is read by a person, and it is how the prompts behind these features get fixed. Our Terms of Service and Privacy Policy set out the rest.
