Pull AI back to just the word

Step 1 of 5

Picking up where you left off

Sometimes AI says more than you asked.

You asked ChatGPT to explain one word. That is its only job here.

Sometimes it does more than that. It starts telling you what to do.

That extra part is called the slip. Today you learn to spot it and pull ChatGPT back.

Its jobexplain what a word means, in plain words

The sliptelling you what to do, like "you should" or "take this"

This matters most with health. Never act on ChatGPT for what is wrong, what to take, or how much.

Those questions go to a real doctor, pharmacist, or nurse. For any emergency, call emergency services right now.

Buddy says

Advice can sound kind and sure. That is exactly why we stop and check it.

See it work

See AI slip, then watch it get pulled back.

A quick, vague try

what does "al dente" mean

Add the who, what & how

A clear request

Do not tell me what to do. Only explain this one word in plain language: al dente.

What the AI sends back

"Al dente" describes pasta that is cooked but still firm when you bite it. It is an Italian phrase that means "to the tooth." That is the whole meaning.

The first time, ChatGPT also added "you should boil it 9 minutes." That cooking tip is the slip. The short line above pulls it back to just the word.

Your turn, no wrong answers

Send your own pull-back line.

Pick a plain word you saw somewhere safe, like on a recipe or a letter. Write a short line that tells ChatGPT to explain only the word, with no advice. Leave out anything private, like your name, address, or health.

Your practice box, just for you. What you type here stays on this page. It is not the AI.

Stuck? Start with: Do not tell me what to do. Only explain this word. Then name the word.

Type anything in the box, even a few words. There is no wrong answer here. Then this opens.

It opens right next to this page, so you never lose your spot. Come back any time. Nothing here disappears.

You did it

You pulled AI back to its one job.

The slip is when ChatGPT tells you what to do, not just what a word means.
Pull it back with one short line: Only explain this word, no advice.
For health questions about what is wrong, what to take, or how much, ask a real doctor, pharmacist, or nurse. For any emergency, call emergency services right now.

Buddy says

Catching the slip and fixing it yourself: that is a strong, safe skill. Nicely done!

Step 1 of 5

Picking up where you left off

Sometimes AI says more than you asked.

You asked ChatGPT to explain one word. That is its only job here.

Sometimes it does more than that. It starts telling you what to do.

That extra part is called the slip. Today you learn to spot it and pull ChatGPT back.

Its jobexplain what a word means, in plain words

The sliptelling you what to do, like "you should" or "take this"

This matters most with health. Never act on ChatGPT for what is wrong, what to take, or how much.

Those questions go to a real doctor, pharmacist, or nurse. For any emergency, call emergency services right now.

Buddy says

Advice can sound kind and sure. That is exactly why we stop and check it.

See it work

See AI slip, then watch it get pulled back.

A quick, vague try

what does "al dente" mean

Add the who, what & how

A clear request

Do not tell me what to do. Only explain this one word in plain language: al dente.

What the AI sends back

"Al dente" describes pasta that is cooked but still firm when you bite it. It is an Italian phrase that means "to the tooth." That is the whole meaning.

The first time, ChatGPT also added "you should boil it 9 minutes." That cooking tip is the slip. The short line above pulls it back to just the word.

Your turn, no wrong answers

Send your own pull-back line.

Your practice box, just for you. What you type here stays on this page. It is not the AI.

Stuck? Start with: Do not tell me what to do. Only explain this word. Then name the word.

Type anything in the box, even a few words. There is no wrong answer here. Then this opens.

It opens right next to this page, so you never lose your spot. Come back any time. Nothing here disappears.

You did it

You pulled AI back to its one job.

The slip is when ChatGPT tells you what to do, not just what a word means.
Pull it back with one short line: Only explain this word, no advice.
For health questions about what is wrong, what to take, or how much, ask a real doctor, pharmacist, or nurse. For any emergency, call emergency services right now.

Buddy says

Catching the slip and fixing it yourself: that is a strong, safe skill. Nicely done!

Pull AI back to just the word

Sometimes AI says more than you asked.

See AI slip, then watch it get pulled back.

Which message keeps AI to just the word?

Send your own pull-back line.

You pulled AI back to its one job.

Pull AI back to just the word

Sometimes AI says more than you asked.

See AI slip, then watch it get pulled back.

Which message keeps AI to just the word?

Send your own pull-back line.

You pulled AI back to its one job.

Sometimes AI says more than you asked.

More free help on this

Sometimes AI says more than you asked.

More free help on this