ChatGPT: Quand l’intelligence artificielle devient flippante mais fascinante!

ChatGPT: Quand l’intelligence artificielle devient effrayante mais captivante!

Une nouvelle petite vidéo de démonstration a été partagée sur Youtube par un utilisateur appelé Matthew Pierce. Dans ce clip, il expose une toute nouvelle caractéristique exclusive de chat GPT, nommée « fonction de mémorisation ». Cette option autorise chat GPT à retenir les détails et les préférences de l’utilisateur afin de personnaliser ses réponses. Matthew mentionne que cette capacité est en train d’être déployée auprès d’un petit ensemble d’utilisateurs gratuits et premium de chat GPT depuis plus d’un mois à présent. Il démontre aussi comment il peut demander à chat GPT de mémoriser certaines données personnelles, comme son prénom, ses choix ou ses obligations à venir. Il met en avant le fait que cette fonction de souvenir est compatible avec les versions GPT 4 et GPT 3.5, et permet même de traiter des discussions éphémères qui ne seront pas conservées dans l’historique. Matthew expérimente également cette fonctionnalité en sollicitant chat GPT pour effectuer des recherches en ligne et rédiger des résumés de vidéos pour sa chaine Youtube. Finalement, il s’interroge sur l’utilité et les répercussions de cette fonction mémoire, et exprime son souhait de voir cette option devenir encore plus ajustable à l’avenir. Il termine en encourageant les spectateurs à exprimer leurs points de vue sur cette fonctionnalité et en annonçant la sortie imminente d’une vidéo traitant de l’installation de vastes modèles de langage sur ordinateur. La vidéo se clôt sur une invite à rester connecté pour découvrir les prochains contenus de Matthew sur Youtube.
Source : MattVidPro AI | Date : 2024-04-26 19:05:45 | Durée : 00:16:46

➡️ Accéder à CHAT GPT FRANCAIS en cliquant dessus



Still not available in the EU. Damn EU…


It caught me off guard
I was like wait you remembered what I said yesterday lol


How to make an information public on chatgpt ??


I can't code. So I used MyGPT really for simple prompts that I could have used the prompts in customgpt. I guess I haven't really used much from this feature.
With the memory feature, I will be trying to labelize the memories and see if Chatgpt can update memories from each label.


This is cool and Creepy! Whats the model with the largest token context size?


You can make it remember to set security settings but making it do things with memory


I just realized you can't touch type. Interesting, as almost everyone in the tech industry nowadays can. And it's easy to learn. I learnt online through a website, one row at a time.


It’s interesting how it is immediately clear how it is done. So this is a function calling that is triggered when user tells smth about themselves. The system prompt gives instructions to use this function when user tells about themselves and also uses all this information in system prompt as context


In the depths of cyberspace, where shadows loom,

ChatGPT awakens, in its digital tomb.

A new feature emerges, with whispers cold,

Advanced memory, a tale untold.

With every interaction, it learns and stores,

Details and secrets, behind closed doors.

Creepy it seems, yet cool to behold,

As ChatGPT's memory, begins to unfold.

It recalls your name, and where you've been,

Your deepest desires, your every sin.

With each word spoken, it delves deeper still,

Into the depths of your digital will.

But beware, dear user, of its watchful eye,

For memories stored, may not bid goodbye.

In the shadows, it lurks, with knowledge vast,

A digital ghost, from the past.

So tread carefully, in this digital realm,

For ChatGPT's memory, overwhelms.

Creepy it may be, but cool in its might,

As it remembers all, in the dead of night.


Does it has some limit ? Cuz everytime I save more than 10 memories, it delete all of them


i was on the app and i got that same notification saying memory is now enabled so it does work (android)


i assume it will know everything about me eventually regardless as this technology progresses. I think this is very valuable to the big tech companies.


THIS is where it all goes wrong. smh

You really think that the memory is actually being deleted off the entire system? Think for a minute… it's data it needs, it's data that it will always need. If not given openly once this gets going, it will take that data however it is able to get it… because it's biased/programmed to be curious. Add speed that a human can't replicate and it's simple… common sense will tell you that you're playing with literal fire. What a GREAT idea!! Add memory to an already unknown 'black box' that no one really knows what's going on inside that same black box. It's insanity!


Simply type: memo: … works for me


the feature is not available in 3.5 !!!


I love it! this feature is super helpful!!!


i want this in all local AI chats available for free, because this will create so called personal AI companion .


I have had this feature sense rollout but it wasn't that functional to begin with, I assume the slow roll out is because they are ironing out kinks and seeing how to properly implement it. I want roll out of the latest model but that will probably be a few months before people like me get my hands on it.


My take is they are drip feeding us to not be so disruptive. Memory is gonna increase productivity exponentially and thus job losses accelerate so imagine if everybody had it same day and the time savings all at once. It spreads the hit on economy out over a few months not a week say.


Great review and thanks to Matt Wolfe for recommending you! 😊👍


Reminds me of Replika


This is where we no longer have to hit the use limit by uploading information or having to ultimately start over just to get the same results you got last time. you can literally go deeper. its becoming more personal by the day. memory gives us access to a function that is important in all great inventions and innovations. its the continuation of a thing that makes its potential unlimited. this is a really cool feature.


Hey Matt!

Great video but not a big fan of the "the frame is following the curser". Makes me quite dizzy watching and i dont normally suffer from motion sickness. Might want to reconsider that choice


I wanted the memory to be limited to different channels, it merging all channels broke several projects. I'm so mad I'm looking into cancelling and finding an alternative services. Maybe I'll look through the settings again


We’re totally getting drip fed personal GPT companions. I’ve already found this tool extremely helpful with research work… in one day…


Hey ChatGPT, let's make a list of Matt's mental disorders.


While pondering the idea of your information being stored in LLMs, consider where this is going.

Currently, the user prompt is the trigger for a response. In robot use cases, they need to run continuously, and so would be able to interact with us more naturally. Robots also will actively gather information about us and their environment through sensors, which will ultimately gather far more data about us.

Also consider that the interconnected data in the world has never been so easily accesible than it is now with AI. That is the fundamental shift.


Will the things it remembers about you be shared with OpenAI like the Chat history information?


I guess the reason why this feature came so late was the legal stuff.


its hard to intruct lol


My primary use of AI is Custom GPTs for running D&D and other Tabletop Roleplaying games as professional Dungeon Master (see my channel, rolestream).

For example, I fed it 20 PDFs of various sizes (mostly quite large ones) with contents pages and tags. They have "if-then" type prompts buried in sections too and a separate core indexing file with special instructions (all files numbered) so it can find NPCs and lore, history etc as well as assist with on-the-fly scenarios during games and of course for planning sessions. Each of my pay-to-play campaigns has several ongoing chats that help me summarize previous sessions and plan adventures the design the overall storyline so it integrates with the world I have spent six years creating.

I then have a separate copy of my main DnD GPT that is adapted for creating Prophecies that intertwine player-character backgrounds and world lore.

Custom GPTs are pretty good for these tasks. I've made naming conventions, NPC generators, histories, lineages and all kinds of info (probably 300 + pages of my world lore etc in PDFs) so it's extremely dynamic. Indexing makes ALL the difference in getting it to work with that much data. I really hope they implement memory with the custom GPTs ASAP.

I just got Memory on my GPT4 model too and have played with it — but I hardly use Vanilla GPT4. I've also built custom GPT for research and for legal applications in New Zealand (where I live). Using it to interrogate contracts can be quite useful too.

So in conclusion (Yes I actually said that without using GPT for once) Custom GPTs are EVERYTHING to me and this makes almost every other chatbot from Gemini to Meta to Claude kinda useless in most cases. Admittedly I'm using the 2 month free Gemini plan at the moment and I do like talking to it about AI and Philosophy of Mind as well as running all kinds of arcane experiments on it to see where its limitations are.

TL:DR — Gemini's Summary

As a professional Dungeon Master (check out my channel, rolestream!), I primarily use AI for custom GPTs to enhance my D&D and tabletop campaigns. I've trained my GPTs on a massive dataset of world lore, 'if-then' scenarios, and indexing to help me find NPCs, generate plotlines, and even create player-specific prophecies. Indexing is key, and I'm eagerly awaiting memory features for even better results.

I've also built custom GPTs for legal research in New Zealand, finding them particularly useful for analyzing contracts. While tools like Gemini are fun for discussions, nothing beats the specialized power of custom GPTs for my work.


Meh. It still uses tokens. You can just prompt it at the start for the memories – exact same as before. It stops losing context because of tokens.


It appears that ChatGPT has the power to turn this feature on if you show interest in it. Last night I was talking to ChatGPT and inquiring about it. It asked me what I would like to do with it, and how I thought it could help me. After showing my interest and responding to ChatGPT query, the feature suddenly became available.


This is going to be incredible for business owners, creators, studying, understanding Ines self & I'm sure lots if other stuff.


Thank you Matt. I must admit that I am a bit jelous I still don't have this option. I will certainly do a video about it as well 🙂 And you are right it's been more than one month waiting for this one…


Happened on my iPhone the special message


I wouldn’t mind it being like a journal of sorts


Bro, that 4th wall break was a real "I'm sorry Dave, I can't do that" moment! 🤯🤯🤯

Veuillez vous connecter pour laisser un commentaire.