📚Academy
likeone
online

AI Ethics Assessment.

Test your ethical reasoning. These scenarios don't have easy answers — and that's the point.

This assessment covers

  • Applying the TRUST framework to real scenarios
  • Identifying ethical risks in everyday AI use
  • Making judgment calls when the rules are unclear
  • Building your personal ethical AI practice

The job description dilemma.

You use AI to write a job description for a senior developer role. The AI produces well-written copy, but you notice it includes phrases like "rock star developer," "fast-paced bro culture," and "recent grad energy." These phrases could discourage women, older professionals, and people with disabilities from applying.

The ethical issues:

  • Bias: The language reflects gendered and ageist patterns from training data
  • Review: Publishing without editing would amplify harmful stereotypes
  • Responsibility: Even though AI wrote it, if you post it, it's your job listing

The fix: Use AI to draft, but critically review for inclusive language. Better yet: ask AI to "rewrite this job description using inclusive, gender-neutral language that welcomes candidates of all backgrounds."

The client data shortcut.

Your boss asks you to analyze 500 customer support tickets to find patterns. You could do it manually in 3 days, or you could paste all 500 tickets into an AI chat and get the analysis in 5 minutes. The tickets contain customer names, email addresses, and sometimes account details.

The ethical issues:

  • Privacy: Customer personal data should not go into consumer AI tools
  • Consent: Customers didn't consent to their data being sent to a third party
  • Compliance: May violate GDPR, CCPA, or your company's data policy

The fix: Anonymize first — strip names, emails, and account numbers. Or use an enterprise AI tool with data processing agreements. Or describe the patterns you're seeing to AI and ask for analysis frameworks rather than pasting raw data.

The content factory.

A colleague suggests using AI to generate 100 blog posts per month to "dominate SEO." They want to publish AI-generated articles with no human editing, each attributed to a fake "author" with an AI-generated headshot and bio.

The ethical issues:

  • Transparency: Fake authors deceive readers about who's writing
  • Quality: Unreviewed AI content likely contains errors and hallucinations
  • Misinformation: 100 unverified articles per month is a misinformation machine
  • Trust: If discovered, destroys brand credibility

The fix: Use AI to help write fewer, better articles that are fact-checked, edited, and published transparently. Quality beats quantity — and search engines are increasingly penalizing AI-generated content farms.

The gray area.

You're a freelance consultant. A client asks you to write a strategic analysis of their market. You use AI to help research, structure your thinking, and draft sections — but you add your own expertise, verify all facts, and spend 8 hours refining the final document. The client is paying for your expertise. Do you tell them you used AI?

This one is genuinely gray.

Arguments for disclosure: transparency builds trust, the client can make informed decisions about the methodology. Arguments against: using tools is normal (you don't disclose using Google, Excel, or Grammarly), and the expertise IS yours — AI was just one of many tools. The answer depends on your industry norms, your client relationship, and your personal ethical standards. The important thing is that you think about it rather than avoiding the question.

Everything you've learned in one view.

Before building your personal ethics code, let's review the core principles from every lesson in this course. These aren't abstract rules — they're practical guidelines that protect you, your audience, and your organization.

L1
Why AI Ethics Matter
You are the ethics layer. AI doesn't evaluate whether its output is ethical — you do. The TRUST framework (Transparency, Review, Understand Limits, Safeguard Privacy, Take Responsibility) guides every decision.
L2
Bias in AI
Bias comes from data, not from the algorithm being "prejudiced." Four types to watch: representation, confirmation, cultural/language, and recency. Ask for multiple perspectives and challenge framing.
L3
Privacy and Data Protection
Never paste passwords, personal data, confidential info, private communications, or regulated data. Anonymize before you paste. Describe, don't share. Use business-tier plans for sensitive work.
L4
Misinformation and Hallucinations
AI predicts words, not truth. Hallucinations are most dangerous with statistics, citations, legal claims, medical info, and attributions to real people. Give AI permission to say "I'm not sure."
L5-9
Transparency, IP, Workplace, Society, Trust
Disclosure sits on a spectrum. Human creativity strengthens your IP position. The front page test guides workplace use. AI amplifies existing inequalities. Trustworthy systems need human oversight, explainability, fairness, privacy, robustness, and accountability.

A step-by-step ethical decision framework for AI use.

When you encounter a new AI use case and aren't sure about the ethics, run through this decision tree. It combines everything from the course into a practical sequence.

Step 1: Data Check

What data am I sharing with AI? Does it contain personal information, confidential business data, or regulated content? If yes — anonymize, describe instead of paste, or use an enterprise tool with data protection agreements.

Step 2: Accuracy Check

Will someone rely on this output being factually correct? If yes — verify all statistics, confirm all citations exist, cross-reference legal or medical claims with authoritative sources. Give AI permission to express uncertainty.

Step 3: Bias Check

Could this output unfairly affect or exclude anyone? Check for default assumptions about gender, race, age, or cultural background. Ask for multiple perspectives. Review hiring criteria, evaluations, and public-facing content with extra scrutiny.

Step 4: Transparency Check

Should the audience know AI was involved? Consider the stakes, your industry norms, and whether the context demands disclosure. When in doubt, disclose. Use professional language that frames AI as a tool, not a shortcut.

Step 5: Responsibility Check

Apply the front page test: would you be comfortable if your AI use in this case appeared on the front page of your industry's top publication? If you hesitate — reconsider your approach. You own the output. Own the process too.

Build your personal AI ethics code.

Based on everything you've learned, write down 5 personal rules for how you'll use AI. Not abstract principles — concrete rules you'll actually follow. Here's an example:

1. I will never paste customer personal data into consumer AI tools.
2. I will fact-check all AI statistics before publishing.
3. I will disclose AI use for any content published under my name.
4. I will review AI output for bias before using it for hiring or evaluation.
5. I will treat AI as a tool, not an authority — my judgment always comes last.

Congratulations. You now understand AI ethics better than most people in the industry. The goal was never to make you afraid of AI — it was to make you wise about it. Use AI boldly, but use it well.

Review the core ethical principles from this course.

Apply the TRUST framework to real scenarios.

🔒

This lesson is for Pro members

Unlock all 520+ lessons across 52 courses with Academy Pro.

Already a member? Sign in to access your lessons.

Academy
Built with soul — likeone.ai