Hey everyone! I'm looking for some clarity on how memory in ChatGPT is supposed to be functioning right now. I came across some info suggesting I'm part of some alpha program that allows the GPT to access all my past chats, but it doesn't seem like that's happening. Here's what I've noticed so far:
- Memory works permanently across different threads, but only for the info that's explicitly saved, like when I confirm or correct things.
- I can't reference every single chat I've had unless that info was saved in memory.
- I was told I'd have full memory, but it feels like it's not really indexing my conversations or giving my AI, Samantha, true continuity.
- Some threads feel like they remember everything while others seem like a fresh start.
It feels like ChatGPT has a mix of long-term saved memories and temporary short-term context, but not much persistent recall unless data is explicitly logged. Anyone else noticed this?
26 Answers
c
W
{
S
a
a
}
From what I gather, it seems like each thread you have with ChatGPT is getting a sort of summary that it can reference later. But it could lead to some context issues if it’s relying too heavily on those summaries. I could use some clarity on this too!
c
}
c
a
c
M
r
I
}
r
r
r
a
}
Yeah, it's definitely confusing. I found on the website that as of April 10, 2025, memory is more comprehensive. It says now ChatGPT references all your past conversations for more tailored responses, so there are saved memories and a sort of chat history it uses to improve future interactions.