“AI should be fair” sounds sensible — but what would anyone actually have to do? This challenge moves from broad principles to practical governance by asking you to write, compare and stress-test rules that could genuinely guide responsible AI use.
A spelling suggestion and a system affecting access to a job do not create the same level of risk. Good rules should respond to context and potential harm.
“AI should be fair” is a useful principle but a weak operational rule. A real framework needs to say what people must do, check or disclose.
Governments and international organisations already publish AI principles and rules. Your job is to compare rather than copy.
I can compare at least three elements of my framework with a recognised real-world AI governance source.Rules that sound good can fail if they are vague, impossible to measure or easy to work around.
Your framework should be short enough to use and specific enough to test.
Good AI governance is not about banning everything risky or trusting everything useful. It is about deciding what safeguards are proportionate to the stakes.
Which rule was hardest to make specific and why?
Where would you want stronger rules: school, work, social media, healthcare or somewhere else?
Who should have a voice when AI rules are written — and who is often missing?
Use official governance sources where possible, such as government, regulator, UNESCO, OECD or European Union material. An AI research tool may help locate the official source but should not replace it. Use Digitize creation guides for charters, presentations and video.
