I've been encountering a strange issue with my GPT where it delivers seemingly random responses that don't relate to the current conversation, although they seem to be drawn from past interactions I've had with it. The chat works fine in temporary mode, and short texts elicit normal replies. However, when I send longer messages or ones that include non-text content like images, it tends to respond randomly. My GPT memory is also full since my subscription expired, but deleting memories hasn't freed up space, and I value the remaining information. I've shared examples of this behavior in some screenshots, illustrating how the same question yields different responses in regular versus temporary chats. Given that temporary chats work correctly, I suspect the issue might be tied to memory management. Is anyone able to explain what's going on or suggest a fix? I've noticed others mentioning similar problems after their subscriptions lapsed, which makes it seem like it might be a tactic to push for Plus subscriptions. What do you think?
2 Answers
I've had the same issue! It seems like the bot can get a little lost if you dive deep into a topic. It's like it has a bit of cognitive overload and starts mixing things up. I found that switching to a new chat helps keep responses relevant. Just don’t go too deep too quickly, and it should behave better.
It sounds like your GPT is struggling with managing its memory after so many interactions. If the memory feature is active, it might be holding on to too much old info about the same topics, causing confusion. I recommend going through and pruning those memories, especially when you bring up new subjects—try starting fresh in a new chat for different topics.
True! Sometimes it feels like it’s trying to dodge certain topics because the context is jumbled.