ChatGPT Memory and Chat History Usage Practicalities

A few days before the announcement of the new feature for referencing past conversations I deleted all my chat history… I did this because I was tired of trying to clean up thousands of past chats many of which were just small stubs where I had to keep creating fresh chats to restart and not have polluted conversations that kept referencing things I wasn’t talking about…

This has changed more recently and now often chats don’t reference when I want them to.

This aside I am now considering how to structure my saving of chats for Memory and Chat History reference…

From research I have done so far I understand that:

OpenAI hasn’t revealed exactly what algorithms they use to index your chats so it’s very hard to understand what biases might exist.

Here is a little research I did on 4o:

🧠 How Relevance Is Determined
  1. Semantic Matching
    Instead of looking for exact words, ChatGPT uses embeddings (numerical representations of meaning) to understand the essence of your past interactions. It compares those embeddings with your current prompt to see what aligns best.
  2. Recency Bias
    Recent interactions are usually considered more relevant—so if you talked about a trip yesterday and bring it up again today, that recent context is prioritized over something from months ago.
  3. Frequency and Importance
    If a detail (like your profession or favorite writing style) comes up a lot, or you’ve asked ChatGPT to remember it explicitly, it’s flagged as more important and thus more relevant to future replies.
  4. Conversation Intent
    ChatGPT tries to infer your current intent. For example, if you’re asking about design trends, it’ll pull relevant past design-related chats or memories, not your preferences for coffee (unless you bring them up in a related way).
  5. Confidence and Fit
    If there’s low confidence that a memory or previous detail is useful in the current conversation, it might not use it. Relevance isn’t just a match—it also depends on how useful a detail is for improving the response.
🎮 Ways to Game Relevance in ChatGPT

1. Repetition = Importance

  • Mention something multiple times (e.g. “I’m a freelance UX designer, so I always care about clean interface design”) across different chats.
  • Repeating key facts nudges ChatGPT to treat them as core parts of your identity or preferences.

2. Use Specific Language

  • Phrasing like “This is important to me,” or “Always assume this,” can help flag concepts as high-priority.
  • Example: “I prefer bullet points over paragraphs when skimming ideas—please keep that in mind going forward.”

3. Establish Context Early

  • At the start of a conversation, lay out key points you want ChatGPT to frame its answers around.
  • E.g., “For this whole chat, assume I’m building a startup aimed at eco-conscious Gen Z users.”

4. Set Traps (in a good way)

  • Mention contrasting facts and see which one it picks up in future chats to test what it remembered or found more relevant.
  • You can say: “I’m not sure if I prefer minimalism or maximalism in branding. I lean minimalist but love bold color.” Later, see what it assumes.

5. Use Memory Intentionally

  • If memory is on (in supported regions), explicitly ask it to “remember” things you want it to prioritize long-term.
  • You can say: “Remember that I’m an intermediate Python developer who prefers real-world code examples over theory.”

6. Trigger with Keywords

  • Since relevance often hinges on semantic similarity, using similar phrasing to your past prompts can help retrieve related ideas.
  • Think of it like giving the system breadcrumbs to follow.
🧮 How Many Data Points Can ChatGPT Consider?

1. Context Window Limits (per chat)

  • Models like GPT-4-turbo (what you’re chatting with now) can handle up to 128,000 tokens of context (~300 pages of text).
  • That includes your input, ChatGPT’s responses, and any relevant stored memory or chat history it pulls in.
  • It doesn’t consider everything ever said—just what fits in that token window, prioritizing relevant and recent information.

2. Memory Recall Scope

  • For long-term memory (the saved “you-asked-me-to-remember” kind), it doesn’t need to load all data at once.
  • Instead, it likely uses a semantic index to retrieve only the top N most relevant entries—think: the top 5–20, depending on context.
  • These are picked by comparing embeddings (aka: meaning) between your current input and stored memories.

3. Chunking and Retrieval

  • Larger or older pieces of data are “chunked” into smaller bits.
  • When you ask a question, ChatGPT semantically matches your prompt to these chunks and pulls in the closest ones. Like vector search in action.

If anyone has had someone use their ChatGPT account randomly (my son used mine once to edit some code) then they will know that memories saved even from a brief usage can heavily influence future chats if memories are saved…

I have spent some time trying to manage Memories and now see an opportunity to finally get it right with this new feature…

Concerns I have are for example like in the “:video_game: Ways to Game Relevance in ChatGPT” dropdown I’m worried that I will convince ChatGPT to focus on the wrong thing…

ie If someone used my machine and said:
“Create a memory ‘It’s important to always focus on long outputs’”
… without going through all memories I would likely not know it was there…

While this is an extension of self it is so easy for someone to ‘install a virus’ in your thought processes.

I’m assuming that Memories are probably better to guide how chats should be indexed?

I am really interested in anything others have to share outside of the information in the docs.

I will think on this more deeply over the next few days and try and tidy up my topic here a bit more but I think this is an important topic for discussion outside of the current ‘Announcement’ thread

1 Like

Follow up post to keep things tidy ^^

OpenAI References

Customize ChatGPT

Memory Customisation

Some important points

From the quotes below and with a little assumption there appears to be some Hierarchy to Memory…

I included 2 details panels above ‘Customize ChatGPT’ (Settings rather than ‘Memory’ but about same difference) and ‘Memory Customisation’ (ChatGPT Memory and Chat History Memory)

Hierarchy

Customize ChatGPT Settings - Always Considered Every Message

ChatGPT Memory - Always Considered (If Relevant To Prompt) Every Message

Chat History Memory - Some Details Considered (If Relevant To Prompt) Every Message

** Backup and Restore **

Hierarchy

Customize ChatGPT Settings - Manual

ChatGPT Memory - I think Select/Copy/Paste and then individually add after?

Chat History Memory - There is a backup option for chats, sent to email but No Restore afaik? (Though you can archive chats)

Deletion Delay

Open Questions I cant find reference to

Determining Relevance
Ranking Memories ie Hierarchical Method or marking as ‘Important’

1 Like

Thank you for your response, very interested in how can you apply this specifically to the options on OpenAI Memory related options on ChatGPT?

from the FAQ:

When “Reference chat history” is on, ChatGPT will keep information from past conversations unless you delete (or archive) those chats.

I wish it would be possible to recall also from archived conversations.

1 Like

Hi you are trying to reply with AI, and it’s really not up to the task yet without a lot of human guidance.

Use ChatGPT to translate maybe to format your ideas but asking it to reply to messages is way beyond it’s current scope

Unless an option is added to quickly identify ‘stubs’ like this and delete them it’s impossible to keep them in check…

Some kind of similarity groupings and group deletions would be useful even one click deletion on chat.

I assume the 3 click delete on each chat is so people don’t accidently delete their virtual friends…?

This is another thing… There seems to be at least dual purpose here on the same interface… Some having virtual friends and others working on their subscription… You need multiple accounts to do both I guess?

This conflates memory usage without more management options.

I cleared my chat history days ago but already I have 30 stubs+/day to slowly go through deleting :confused: . Once a ‘Temporary’ chat is started I see no way of making it save to history again so it’s not a decision you can make if it works…

I was just watching this:

Sam Altman: OpenAI's Sam Altman talks ChatGPT, AI agents and superintelligence — live at TED2025 | TED Talk.

It got me wondering when I would be able to include the mountain of old hard drives I have stacked up over the years, including family photos and thousands of coding scripts I have written.

Still… Continuing to update this post at the top of the thread: ChatGPT Memory and Chat History Usage Practicalities - #2 by phyde1001 with new information as I find it.

I’m not sold on the whole ‘create your daemon/avatar/virtual self’ idea however I can see how useful this feature might be in this use case… Seen a lot of complaints that on the depreciation of a model that these ‘entities’ are no longer who they were, how will they grow over different models with memory?

On the flipside, for those who use ChatGPT as a technical tool… Should you recreate your chats with each model? You’re not comparing like for like… Dall-E vs 4o Imagegen, buggy script outputs vs much more layered technically accurate outputs, the processes just aren’t the same…

Whatever you do… Has your understanding of what’s possible combined with far more sophisticated models meant that basing what you are doing on old ideas is not an efficient way forward? Models are changing so fast is looking at old chats like this?

If anyone finds new information / use cases / examples then post them here. :slight_smile:

Another thought…

How do you / should you enter information into chats from other sources to get it all referenced up together or wait for OpenAI to add new features for different file types / services…?