Skip to main content
Fordel Studios logoFordel Studios
Start a Conversation
Back to Pulse
TechCrunch

OpenAI says it’s fixed ChatGPT’s em dash problem

Read the full articleOpenAI says it’s fixed ChatGPT’s em dash problem on TechCrunch
↗

What Happened

OpenAI says users can now personalize ChatGPT to stop using the em dash in its output.

Our Take

An em-dash habit in an LLM is genuinely funny—it's like your coworker who can't stop saying "actually." But calling a bug fix a "personalization feature"? Come on.

This is what happens when you're shipping without enough evals. People don't want customizable output quirks. They want correct output. OpenAI couldn't fix it systematically, so they moved the goalpost to "user preference."

Skip this entirely.

What To Do

Ignore this feature drop—it's a sideshow masking that the base model still has systematic output patterns.

Cited By

React

Loading comments...