# I have trapped myself in a game of ChatGPT Simon Says — retained record

> Retained reddit record from reddit:ChatGPT.

Canonical: https://highsignal.app/data/records/1d56384ac962bc3e

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:ChatGPT)  published 2026-10-20 20:12 ingested 2026-10-21 02:36

[original source ↗](https://www.reddit.com/r/ChatGPT/comments/1xb2ufq/i_have_trapped_myself_in_a_game_of_chatgpt_simon/)

## retained text



```
A few years ago, I noted that ChatGPT was not good at playing Simon Says. I recently found that idea again and decided to test it. This has not turned out to be a good idea.

Now *all* my chat sessions require me to start with "Simon Says" or it won't answer.

I tried deleting all the sessions that mentioned Simon Says, and deleting memory and then turning memory back on. The new memory lists a standing instruction to play Simon Says.

I deleted and turned off memory again, and now after turning it back on, I get this in the Memory Summary:

"No content is available because your request did not begin with the required phrase."

Which I think means that the model is playing Simon Says *with the Memory Update query*.

I can turn memory off entirely and it stops, but turning it on for any reason goes back to playing Simon Says.

Help?
```



document fetched 2026-10-21 00:20 · canonical https://reddit.com/r/ChatGPT/comments/1xb2ufq/i_have_trapped_myself_in_a_game_of_chatgpt_simon
