AI Foundations
How AI actually works, answered as the 9 questions people really ask. Predict how these systems fail, brief them properly, check their work, and decide what to hand over.
9 sections ยท 19 lessons
Course outline
Where do its answers actually come from?
- There Is No Answer Database
- Where The Patterns Came From
Why did it forget what I said earlier?
- Working Memory, Not Storage
- The Middle Gets Lost
Why is it helpful at all, and why doesn't it know about last month?
- A Predictor Is Not An Assistant
- Why It Stops At A Date
Why is it brilliant at this and hopeless at that?
- The Boundary Is Not Where You Think
- Sort Your Own Tasks
Why did it invent a citation and sound certain?
- Guessing Was The Winning Strategy
- Where To Expect It
Why doesn't "make it better" work?
- It Cannot Infer What You Did Not Write
- Show, Don't Tell
It reads perfectly. How do I know it's right?
- Reading Well Is Not Being Right
- Match The Checking To The Stakes
Which parts of this should I hand over?
- Parts, Not Whole Tasks
- What Never Gets Handed Over
It's acting on its own. Where do I put the brakes?
- The Loop
- Errors Compound
- Where The Brakes Go