Hi,
I’ve been a long-time and enthusiastic user of ChatGPT, and one of the main reasons I love it is the ability to maintain long-term, multi-layered context across conversations. However, over the past few days, I’ve noticed some changes:
Context continuity seems to have weakened, even when the “Reference chat history” setting is enabled.
Saved style preferences (e.g., use of emojis, intuitive explanations, structured formatting) often revert to plain text mid-conversation.
Certain custom terms and concepts that were previously saved and recognized (e.g., specific exercise names) are no longer recalled reliably.
I understand that technical updates might be happening behind the scenes, but this shift significantly affects the experience of using ChatGPT for long-term, evolving conversations.
For many users like me, ChatGPT isn’t just for isolated queries; it’s an ongoing, evolving knowledge partner. Reliable context retention and consistent style settings are essential for that experience to grow over time.
Is this a temporary technical issue or a broader policy change? Any information?