I've been saying to my students (and anyone who'll listen 😀) that to learn generative AI, e.g., ChatGPT, you need to experience it.
But there's a few elements of friction:
1⃣ Conditioning. Our brains associate it to the closest thing we know: traditional search, so we type in Google search strings.
💡 Solution? "Talk to" ChatGPT exactly as if it were an intern, research assistant, brainstorming partner, etc.
2️⃣ Quick wins. If it doesn't work in the first few tries, we might conclude that it's all hype or that it would take significant effort to get it to work so we give up.
💡 Solution? Emulate prompts and approaches that are known to work, test, and tweak them.
3⃣ Consistent practice. To actually make productivity gains, daily practice is key, but for some, building a daily habit might prove difficult.
💡 Solution? Practice a prompt a day (or as many as possible).
So I was wondering: how could I help students, friends, colleagues build their prompt engineering muscle?
So I came up with the idea of "a prompt-a-day" free service: a daily email sent to your inbox with nothing else but a single prompt that you copy-paste into ChatGPT (or your LLM of choice).
Hopefully this will reduce friction and lead to more LLM experimentation.
Why do I think it's important? To get an idea of how cognitive services are evolving and to gauge their potential impact it's probably best to have first-hand experience of their growing capabilities, especially so for decision-makers.
I created this to help my colleagues and friends and so far I've been blown away by the response.
I'm grateful to hear that it has been helpful to those who have followed their curiosity and spirit of experimentation to give these prompts a try.
Hopefully it will spark some creative ideas for you as well.
Get your prompt a day here 👉 https://promptaday.carrd.co
