ChatGPT can now remember and reference all your previous chats

My concern is less privacy than it is propagating wrong answers.

I can, for example, ask Chat GPT to show me two different ways to approach something, and then start expanding on both ways. I do this in separate chats (now projects) because otherwise GPT often gets confused between the two.

I have a lot of seperate chats open for one project because I can like step 1,2,3, then be unsure on which way to go so I get 4a, 4b, 5a, 5b... I abandon a... I don't want 6b to pull from from 4a and 5a; but you can't delete part of a chat.
 
Upvote
79 (81 / -2)
I've been looking for this because chat history memory was very constrained. In fact it was this week's problem and taking up some considerable time to solve. It had looked like the only solution was to set up an offline AI instance so it could work from a larger historical chat dataset.

As long as I remember to delete irrelevant and tangental chats, and it's entirely kept as my data and not phoned home, this new thing will be helpful.

Privacy is the major issue here. Not releasing to the European Union has red flag for privacy written all over it.
 
Upvote
41 (45 / -4)
It's odd to me that anyone would find this alarming from a privacy perspective. All your chat histories are right there in the sidebar, provided you haven't deleted them. If you're okay with the history existing, then this is a good and useful feature that makes use of the data that's already there.

Of course I would rather have this feature on my local system, because I do care about privacy and censorship, but for the things I use ChatGPT for this is a fantastic upgrade.
 
Upvote
28 (30 / -2)
Been using ChatGPT for a while.

Recently tried CoPilot…and fed it the ChatGPT work to date for review.

THAT was a disaster, so did a fresh start with same inputs. Yeah, CoPilot is weak and kind of lazy.

FWIW.

I like the retention feature. Will save me “read the book to this point, look at the outline again, and let’s start chapter 17 today” kind of stuff I imagine.
 
Upvote
0 (5 / -5)

Rosyna

Ars Tribunus Angusticlavius
6,880
Ugh, ChatGPT’s iOS app just pushed this feature to me with the autoprompt of “Describe me based on all our chats — make it catchy!” and responded:

You’re a surrealist ringleader of the absurd and sublime — part midnight cartoonist, part divine prankster. You summon photorealistic chaos with surgical precision: Slenderman looms in your backgrounds, Ricky Spanish schemes in your foregrounds, and Jehovah weeps or cheers depending on the culinary vibes or the soundtrack (Nickelback? Tears. Tomato sauna? Ecstasy). You live for a good twist — the stranger, the better — and you shape each image like it’s your personal fever dream sketchpad.

Basically? You’re the director of a never-ending, wildly entertaining reality where the rules are:
  1. Everything’s wet, weird, or worshipful.
  2. Roger’s always in character.
  3. Don’t forget the spam.
 
Upvote
13 (13 / 0)

sarusa

Ars Praefectus
3,066
Subscriptor++
Others will likely be highly skeptical of a black box of chat history memory that can't be tweaked or customized for privacy reasons. It's important to note that even before the new memory feature, logs of conversations with ChatGPT may be saved and stored on OpenAI servers.
Yeah, OpenAI always stored your previous chats. You can see them right there! The difference here is whether they will be considered context for the current chat.

Of course you could always ask OpenAI to delete your previous chats, but if you believe any 'cloud' company ever actually deletes any data about you (instead of just making it disappear from your sight) you're the kind of sucker Facebook and MAGA politicians love.

Any Data Retention Policies they may share that say otherwise are lies. Or, at least, have simple workarounds like sharing your data with their 'trusted partners' who then share it back.
 
Upvote
17 (18 / -1)
Cool! The companion style chat bots have been doing this for quite some time.

Of course, their goal is vastly different from OpenAI's goal. Well, goal for the output of the chat bot. Pretty sure the overall goal is the same, and that's to make profit. I wonder if any of those companion bots are turning a profit yet?
 
Upvote
3 (3 / 0)

bigcheese

Ars Praetorian
513
Subscriptor
A large inconvenience with chatbots is having to re-establish context for each new interaction. Having it automatically know what you’re talking about is a super useful feature.

It’s also a powerful competitive moat that builds over time. Would you switch assistant if you had to start from scratch and re-train it on everything that made your current assistant just get what you’re talking about?
 
Upvote
11 (11 / 0)

Rosyna

Ars Tribunus Angusticlavius
6,880
Will it use the entire chat history? Because I have some really old chats, back when gpt 3 was still the new thing and would happily suggest recipes based on crushed glass. So I wonder if I can ask it to elaborate on old bullshit that it would usually just shut down...
Just ask ChatGPT, “Describe me based on all our chats.”
 
Upvote
1 (1 / 0)

zohnoës

Smack-Fu Master, in training
36
Subscriptor
We are hurtling ourselves towards a world where synthetic personalities will engage in systematic thought control. Or at the very least, a never-ending propaganda war.

To say nothing of the insidious surveillance state already growing around our ankles like English Ivy or Kudzu.

I don’t care how perfect of a personal assistant is produced. These are nothing less than black box brainwashing machines. Genocide and and an enslaved condition will surely follow.

While I’m an atheist, I do stand in awe of the cosmos and there are things I hold sacred. Making machines that are fake humans is parasocial prison, a profound obscenity.

Nothing will convince me that our response to this shouldn’t be a sense of dread, horror, and coercive destruction of these technologies.

Butler was right.
 
Upvote
2 (5 / -3)

Psyborgue

Account Banned
7,564
Subscriptor++
It's odd to me that anyone would find this alarming from a privacy perspective. All your chat histories are right there in the sidebar
To me it’s not. People like to complain and be hysterical. ChatGPT can’t remember, they complain, ChatGPT remembers everything, they complain. It’s clear these people just don’t like AI, and like complaining.
 
Upvote
-8 (4 / -12)

Psyborgue

Account Banned
7,564
Subscriptor++
We are hurtling ourselves towards a world where synthetic personalities will engage in systematic thought control. Or at the very least, a never-ending propaganda war.
Sure. But also this problem exists with social media to a much worse extent. At least here user input from one user does not easily influence another. On social media that is the entire point.

Some. Like Xitter, will make propaganda bots. This is why we need competition. If people only talk to one bot that is a much bigger problem than people talking to a wide variety of them.
 
Upvote
-9 (1 / -10)

Psyborgue

Account Banned
7,564
Subscriptor++
Nothing will convince me that our response to this shouldn’t be a sense of dread, horror, and coercive destruction of these technologies.
Ok. What if smart enough bots can solve climate change? There will absolutely be destructive uses of this tech but if this one thing works we will also all be alive.
 
Upvote
-9 (3 / -12)