7 · Bias, ethics, and honesty

Because models learn from human-made data, they can absorb and repeat bias: stereotypes, skewed assumptions, gaps for less-represented groups (NIST, 2023). Treat AI output as a draft from a flawed assistant, not a neutral authority.

Use it honestly:

• Disclose AI help when it matters (school, work, journalism) per the rules you're under. Passing AI work off as fully your own can be plagiarism or academic dishonesty.
• Don't use it to deceive or harm: no fake reviews, fake people, impersonation, harassment, or cheating.
• Keep humans accountable. You are responsible for what you publish or act on, even if AI wrote it.
• Watch for bias in hiring, grading, lending, or any decision about people; AI should support human judgment, not replace it.

The goal isn't fear, it's responsible use: get the speed of AI while keeping your integrity.

Sources

• National Institute of Standards and Technology. (2023). AI RMF 1.0, "Fair: with Harmful Bias Managed." https://www.nist.gov/itl/ai-risk-management-framework