It’s not a glamorous example, but that’s kind of the point.

There is a specific kind of danger in asking "what should we do for dinner?" before coffee.

Especially when the answer is:

"I don't care."

Because somehow, by 4pm, everyone cares.

This was me and Alex.

We are not recipe people. We are not trying to make weeknight dinner a lifestyle brand. We have meals we like. We know our go-tos. We have kids to feed. We are not looking to make Tuesday night more aspirational.

The problem was never really "what should we make?"

The problem was that dinner had become a recurring decision we were re-making every single day.

  • What do we have?

  • What are we sick of?

  • What night is going to be chaos?

  • Who is pretending not to care but will care later?

So one day, on a car ride, we recorded ourselves talking through it.

Not a polished prompt.

Not a perfect meal plan.

Just a very normal, slightly annoying, deeply domestic conversation about why dinner kept becoming a thing.

Then I gave the messy conversation to ChatGPT.

And because I had already been building what I call my Chief of Me, basically ChatGPT with context about my actual life, it gave us something weirdly useful.

TLDR: Generic ChatGPT gave meal-planning tips. My Chief of Me helped remove the decision.

Not 47 new recipes. Not a quinoa bowl. Not a beautiful Sunday meal prep plan that would collapse by Tuesday.

It gave us a dinner operating system.

Here is what that actually means:

Instead of deciding from scratch every night, we now have themes for each day of the week.

Each theme has 2 or 3 go-to meals, plus flexible options depending on energy level.

So Wednesday "Asian night" could mean:

  • quick stir fry at home

  • ordering Thai

  • sushi once a month if we want something fun

Saturday is steak night, which we jokingly call "Mastro's" after our favorite steakhouse, even though we are usually making it at home with toddlers running around.

The genius part was not the meals themselves.

We already knew what we liked.

The genius part was removing the decision.

Generic Chat GPT vs. Chief of Me

Generic ChatGPT gives you tips:

  • keep easy meals around

  • make a grocery list

  • plan ahead

Fine.

Useful-ish.

Also obvious.

My Chief of Me understood the actual problem:

I do not want to discuss chicken before coffee.

Alex saying "I don't care" is not actionable data.

We already have meals we like.

We need a weekly rhythm, not new recipes.

The goal is less friction, not a perfect meal plan.

Generic ChatGPT gave advice.

My Chief of Me gave relief.

The Winner: Chief of Me

The loser: Generic Chat GPT Answer

Life-changing is too dramatic.

But “two fewer annoying things happening before 8am” feels exactly right.

We no longer have to:

  1. Think about dinner food before coffee

  2. Get annoyed by the other person’s fake “I don’t care”

That is the kind of AI I care about.

AI that quietly takes one thing off your plate.

Try this.

Give ChatGPT the actual mess.

If there is a recurring decision you are tired of making, try giving ChatGPT the actual mess instead of asking for generic advice.

Copy this:

I am tired of deciding [thing] over and over.

The annoying part is [why it creates friction].

The people/preferences involved are [context].

The usual constraints are [time, money, energy, schedule, kids, partner, work, etc.].

Please turn this into a simple repeatable system that reduces how often I have to think about it.

Do not just ask for a meal plan.

Tell it why meal planning is annoying.

That is usually where the useful answer lives.

So I am curious:

What would you want AI to take off your plate first?

Hit reply and tell me.

  1. Untangle my brain

  2. Plan my week

  3. Write my emails or texts

  4. Reduce household mental load

  5. Make dinner less annoying

  6. Help me make deicisions

  7. Create content without overthinking it

  8. Something else (tell me more!)

  9. All of it (totally acceptable answer)

Sharing is caring.

Share with someone who has too much on their plate.

Copy and paste this url:

Keep Reading