Incorrect title generation since this morning | OpenAI | Page 1

Since this morning, the chatGPT website has been applying strange titles to newly created threads that appear to be from other users of the app. I suspect something has gone wrong with user tokens, and I am receiving data intended for other uses.

Example 1:

Prompt: Rewrite the ending of Naurto shipuden as a what-if scenario.

What if during the showdown with Kagua, it turned out that the Infinite Tsukuyomi Jutsu that Kagua had preformed before a long time ago was still ongoing, and the world of Naruto was a shared dream Kagua created, she reawakened upon realizing that somebody inside her dream world had become so disatisfied with her vision of a perfected reality that they attepted to use the same technique she had to recreate the world again. Only upon completing the technique, rather than Madara causing everyone to fall asleep into the Genjutsu, he actually breaks Kagua's Genjutsu over the world, returning reality to its origional even harsher state, a world that has never known chakara. Faced with the truth of the origional more desperate reality, what do the characters do? What do they fight for? A return to Kagua's dream, the world they've known for genrations? A new, improved dream without violence, under Madera? Or does anybody left during the confrontation fight to live in the outside alien world?

Response:
(Nothing *remotely* to do with a knowledge cutoff)```

Example 2:
```Title: GPT-3.5, sometimes referred to as "GPT-3.5," is a language model developed by OpenAI. It is an advanced version of the GPT-3 mod" (cutoff just like that)

prompt: Design a custom DND magic item, a staff called "The fault in our stars" what does it do? How does it look? Make sure to base the item on  to the idea behind the meaning of the expression, not simply a 'star' themed staff. 

Response: (Details of homebrew magic item for D&D, no mention of GPT models or versions)```

#Incorrect title generation since this morning