TALAMANA · THE AI LITERACY MAP FOR ARCHITECTURE AND DESIGN · Foundations · AGE 11—15 · FACTUAL · DURABLE
Machines get things wrong
A machine gives a wrong answer as fast and neatly as a right one.
The idea
A machine only knows what it was shown. Bad examples, or missing examples, lead to wrong answers, and the machine gives them with full confidence. AI is not magic. It is maths on data. When the data is bad, the answer is wrong, and it arrives just as quickly and neatly as a right one would.
Why it matters
A person who is not sure usually hesitates. A machine does not. You have to do the hesitating for it.
See it in the studio
An AI tool labels a photo of a mosque as "a palace." It is not lying. Its examples never taught it the difference. Would you catch that mistake? Only if you know more than the machine.
Watch for this
Believing an answer because it came fast and looked neat. Machines are wrong just as fast and just as neatly as they are right.
Try it
Ask an AI tool three questions about your own town, about places you know personally. Check every answer. Count the mistakes it made confidently.
Prove it
Give one example of a machine mistake you caught, and explain what the machine was missing.
How it works
Machines fail differently from people. A tired person makes sloppy mistakes. A machine makes confident mistakes that follow its patterns, so they look right at a glance. Later on this map you will learn to test for them on purpose.
What this idea builds on
- A starting idea.
What this idea opens up
Sources
- mt_bPFToj0OhZ (Marble Skill Taxonomy, CC BY-SA 4.0)
- TEC.U
- Elements of AI
Open this idea on the map · The complete map · Logika · RBDS AI Lab, India · revised every edition.