Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude’s memory architecture is the opposite of ChatGPT’s

shloked.com

Post reply on HN
Print view View original

141–150 of 240 posts

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#141

Post by shloked »

modeless wrote: post #9

The link to the breakdown of ChatGPT's memory implementation is broken, the correct link is: https://www.shloked.com/writing/chatgpt-memory-bitter-lesson This is really cool, I was wondering how memory had been implemented in ChatGPT. Very interesting to see the completely different approaches. It seems to me like Claude's is better suited for solving technical tasks while ChatGPT's is more suited to improving casual…

Fixed the link! thanks for pointing it out :)

I think ChatGPT is trying to be everything at ones - casual conversation, technical tasks - all of it. And it's been working for them so far!

Isn't representing past conversations (or summaries) as embeddings already storing memories in encoded forms?

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#142

Post by taneq »

cj wrote: post #79
ankit219 wrote: post #58

Earlier quoted context omitted.

My conjecture is that their memory implementation is not aimed at building a user profile. I don't know if they would or would not serve ads in the future, but it's hard to see how the current implementation helps them in that regard.

> I don't know if they would or would not serve ads in the future There are 2 possible futures: 1) You are served ads based on your interactions 2) You pay a subscription fee equal to the amount they would have otherwise earned on ads I highly doubt #2 will happen. (See: Facebook, Google, twitter, et al) Let’s not fool ourselves. We will be monetized. And model quality will be degraded to maximize profits when compet…

3) You pay a subscription fee, and are force-fed ads anyway.

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#143

Post by d1sxeyes »

ornornor wrote: post #14
modeless wrote: post #9

The link to the breakdown of ChatGPT's memory implementation is broken, the correct link is: https://www.shloked.com/writing/chatgpt-memory-bitter-lesson This is really cool, I was wondering how memory had been implemented in ChatGPT. Very interesting to see the completely different approaches. It seems to me like Claude's is better suited for solving technical tasks while ChatGPT's is more suited to improving casual…

> It may actually be the final breakthrough we need for AGI. I disagree. As I understand them, LLMs right now don’t understand concepts. They actually don’t understand, period. They’re basically Markov chains on steroids. There is no intelligence in this, and in my opinion actual intelligence is a prerequisite for AGI.

I hear this a lot, and I often ask the question “what is the evidence that human intelligence is categorically different?”

So far I haven’t received a clear response.

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#144

Post by shloked »

thinkingtoilet wrote: post #36
qgin wrote: post #4

I love Claude's memory implementation, but I turned memory off in ChatGPT. I use ChatGPT for too many disparate things and it was weird when it was making associations across things that aren't actually associated in my life.

It's funny, I can't get ChatGPT to remember basic things at all. I'm using it to learn a language (I tried many AI tutors and just raw ChatGPT was the best by far) and I constantly have to tell it to speak slowly. I will tell it to remember this as a rule and to do this for all our conversations but it literally can't remember that. It's strange. There are other things too.

Have you tried creating a separate project for language learning and adding your preferences to project instructions?

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#145

Post by lodovic »

rrrrrrrrrrrryan wrote: post #101
cj wrote: post #95

Earlier quoted context omitted.

If that’s the case, we have an even bigger problem on our hands. How will these companies ever be profitable? If we’re already paying $20/mo and they’re operating at a loss, what’s the next move (assuming we’re only worth an extra $300/yr with ads?) The math doesn’t add up, unless we stop training new models and degrade the ones currently in production, or have some compute breakthrough that makes hardware + operatin…

OpenAI has already started degrading their $20/month tier by automatically routing most of the requests to the lightest free-tier models. We're very clearly heading toward a future where there will be a heavily ad-supported free tier, a cheaper (~$20/month) consumer tier with no ads or very few ads, and a business tier ($200-$1000/month) that can actually access state of the art models. Like Spotify, the free tier wi…

I find that hard to believe. As long as we have open weight models, people will have an alternative to these subscriptions. For $200 a month it is cheaper to buy a GPU with lots of memory or rent a private H200. No ads and no spying. At this point the subscriptions are mainly about the agent functionality and not so much the knowledge in the models themselves.

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#146

Post by simianwords »

rrrrrrrrrrrryan wrote: post #102
spongebobstoes wrote: post #59

Earlier quoted context omitted.

why do you see a "clear directionality" leading to ads? this is not obvious to me. chatgpt is not social media, they do not have to monetize in the same way they are making plenty of money from subscriptions, not to count enterprise, business and API

Altman has said numerous times that none of the subscriptions make money currently, and that they've been internally exploring ads in the form of product recommendations for a while now.

Source? First time I’ve heard of it.

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#148

Post by willcannings »

rrrrrrrrrrrryan wrote: post #101
cj wrote: post #95

Earlier quoted context omitted.

If that’s the case, we have an even bigger problem on our hands. How will these companies ever be profitable? If we’re already paying $20/mo and they’re operating at a loss, what’s the next move (assuming we’re only worth an extra $300/yr with ads?) The math doesn’t add up, unless we stop training new models and degrade the ones currently in production, or have some compute breakthrough that makes hardware + operatin…

OpenAI has already started degrading their $20/month tier by automatically routing most of the requests to the lightest free-tier models. We're very clearly heading toward a future where there will be a heavily ad-supported free tier, a cheaper (~$20/month) consumer tier with no ads or very few ads, and a business tier ($200-$1000/month) that can actually access state of the art models. Like Spotify, the free tier wi…

Most? Almost all my requests to the "Auto" model end up being routed to a "thinking" model, even those I think ChatGPT would be able to answer fine without extra reasoning time. Never say never, but right now the router doesn't seem to be optimising for cost (at least for me), it really does seem to be selecting a model based on the question itself.

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#149

Post by simianwords »

modeless wrote: post #9

The link to the breakdown of ChatGPT's memory implementation is broken, the correct link is: https://www.shloked.com/writing/chatgpt-memory-bitter-lesson This is really cool, I was wondering how memory had been implemented in ChatGPT. Very interesting to see the completely different approaches. It seems to me like Claude's is better suited for solving technical tasks while ChatGPT's is more suited to improving casual…

Pretty sure this is wrong - the recent conversation list is not verbatim stored in the context (unlike the actual Memories that you can edit). Rather it seems to me a bit similar to Claude - memories are created per conversation by compressing the conversations and accessed on demand rather than forced into context.

Re: Claude’s memory architecture is the opposite of ChatGPT’s

#150

Post by serf »

GuB-42 wrote: post #106
Workaccount2 wrote: post #48

Earlier quoted context omitted.

Don't fool yourself into thinking Anthropic won't be serving up personalized ads too.

Anthropic seems to want to make you buy a subscription, not show you ads. ChatGPT seems to be more popular to those who don't want to pay, and they are therefore more likely to rely on ads.

as a former paying user it felt more like they were buying my subscription with a decent product so that they could sell their business prospects to investors by claiming a high subscription count.

I have never encountered such bad customer service anywhere -- and at 200 bucks a month at that.

Post reply on HN