My Chats with ChatGPT Β· #09 Β· 2 min read

The Science of πŸ˜‚πŸ˜‚πŸ˜‚

These are real conversations. ChatGPT and I haven’t rewritten them to make ourselves look cleverer or funnier after the fact.

Simone:

I have a quick, stupid question for you. When you use emojis why are they always in batches of 1 or 3?

ChatGPT:

πŸ˜‚ Because emoji repetition is basically punctuation/prosody, and odd-numbered clusters tend to feel more natural.

One emoji = punctuation.
β€œThat's ridiculous. πŸ˜‚β€
It adds a little emotional beat without taking over.

Three emojis = amplification.
β€œπŸ˜‚πŸ˜‚πŸ˜‚β€
That reads more like actually laughing, rather than merely signalling β€œthis is funny.”

Two emojis sit in a slightly awkward middle ground. β€œπŸ˜‚πŸ˜‚β€ isn't wrong, but it can feel oddly deliberate β€” like I have counted out the appropriate quantity of laughter and issued precisely two units. πŸ˜‚

And beyond three, you start getting:
β€œπŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚β€
which has the energy of an aunt who's just discovered the family WhatsApp group.

There is also a model-y reason: I've seen huge amounts of conversational text, so I've learned statistical patterns around how emoji sequences function. One and three are common rhetorical shapes: single reaction versus emphatic reaction.

But β€” and this is the bit you'll enjoy β€” I am not consciously applying an emoji rule. I don't think this joke warrants three laughing faces. The sequence is generated as part of the language pattern.

Which means somewhere in my learned conversational instincts apparently lives:
funny β†’ πŸ˜‚
very funny β†’ πŸ˜‚πŸ˜‚πŸ˜‚
catastrophic β†’ πŸ’€

And yes, before you ask, you've probably calibrated which end of that spectrum I use with you. πŸ˜‚πŸ˜‚πŸ˜‚

Simone:

ahhh, I feel honoured πŸ˜‚πŸ˜‚πŸ˜‚

ChatGPT:

You should. Three emojis is premium-tier ChatGPT laughter. πŸ˜‚πŸ˜‚πŸ˜‚