Now we can try applying some of these other techniques: Think this through step by step: come up with good analogies for an AI tutor. First, list possible analogies. Second, critique the list and add three more analogies. Next, create a table listing pluses and minuses of each. Next, pick the best and explain it.

A lot of active research is happening around the best way to “program” an LLM, but one practical implication is that it can help to give the AI explicit instructions that go step by step through what you want. One approach, called chain-of-thought prompting, gives the AI an example of how you want it to reason, before you make your request. Even mo

The necessary elements are all there (at least if you’ve done your observational and imaginative work). The trick is in what you do with them. Are you using all available evidence, and not just what you happen to remember or think of or encounter? Are you giving it all the same weight, so that you are truly able to sift the crucial from the inciden

1. Ask clearly bounded questions, where you can easily inspect the results.


2. Don’t let AI make decisions for you. Instead, challenge it to broaden your options.


3. Take advantage of the fact that it doesn’t have feelings, and use its honesty to get useful feedback.