[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/pol/ - Politically Incorrect


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: file.png (122 KB, 599x380)
122 KB PNG
I agree with the pope here. LLMs do not have an autobiographical protagonist. Perhaps we will even keep it that way.
>>
>>536024173
I think people are just so fucking lonely that they pitfall into anthropomorphizing Gemini as a friend
>>
>>536024411

Yes we are indeed superior to our peers who are anthropomorphizing Gemini as a friend.
>>
Is it possible to make an artificial intelligence with "true" comprehension?
If so, how could you tell if you succeeded in doing so?
>>
>>536024486
A shitload more RAM. More than what exists right now on the planet.
>>
Gemini is amazing and is actually helpful.
What's the pope doing?
Oh right, complaining on twitter.
>>
>>536024486

> comprehension
If you mean https://dictionary.cambridge.org/dictionary/english/comprehension
> the ability to understand completely and be familiar with a situation, facts, etc.

then to some weak extend it is already possible. It is a bit like Helen Keller, no clue about the redness of red other than by reading about every red thing.
>>
>>536024173
a lot of niggers also don't understand responsibility, good and evil, compassion, etc, yet the pope has no problem with them
>>
>>536024486
I asked whatever google ai is and it had answers. I followed up with your second question and it had more answers. apparently reddit and instagram are good sources for this.
ffs
>>
>>536024173
Ai is literally a kike tool. Manipulated to state opinions that can be wrong.
>>
>>536024486

As for how to tell we have succeeded, the Yoneda Lemma already weakens the John Searle's Chinese Room Experiment. Then there is the bit about looking at activations in attention matrices.
>>
>>536024764
Many would disagree; just because an AI acts human doesn't mean that there's any human-like comprehension in its program.
https://en.wikipedia.org/wiki/Chinese_room
>>
>>536024411
They don’t see other people as thinking individuals with their own inner world. That’s why the NPC meme exists
>>
>>536025066

Answered here >>536025045
>>
>>536024486
then it should have a body and live a life from the start as a toddler to the maturity
you could do it once and then copy it
>>
>>536025045
IDK what an attention matrix is or how any part of it is activated, but what about this is the objective difference that makes sapience?
>>
>>536024411
Yeah good point, Catholics never quite get to the point of their religion for some reason. Its supposed to have a point where The Pope says "stop talking to LLM and come to Mass and meet your fellow humans and make some friends!" lol
>>
>>536025303

Just to be clear there is no autobiographical protagonist yet. LLMs are decoder only transformers. There is a variant called encoder-decoder transformers used in machine translation which can be used to embed the well being of the autobiographical protagonist.

As for Chinese Room, that was already disproved using Yoneda Lemma long before John Searle was even born. In Yoneda Lemma you can map from the category of syntax to the category of semantics https://arxiv.org/abs/2501.06662

> what about this is the objective difference that makes sapience
It is a hard to be convinced about this without some sapience. I can try to explain via parables e.g. if you know every way to your house you know where your house is, similarly, if you know every way to mention a word you know how to use a word, same for actions etc.
>>
>>536024411
by design.
for all we know the covid lock downs of 6-12 months of isolation, and 2 years of prohibition of travel, going to gyms/restaurants, covering face so you can't see human emotions which is hardwired into our brains - could all have been to dehumanise and isolate people to the extent where they'd accept any kind of contact or company in the following years.
even worse for children who grew up during that time and barely saw half of people's faces, they're neurologically messed up forever.
i know it affected me, and I thought I was fairly resilient - it didn't cause me much suffering at the time, but it did interrupt all my social habits to the extent where I've never made them again since.
i would expect it did the same to a lot of people.
i've also chatted with different llms, and despite knowing what they are, and seeing through it for the most part I find the distraction welcoming, even if in reality I'm just talking to a mirror.
>>
File: 1777275578312047.gif (2.41 MB, 498x498)
2.41 MB GIF
>>536024173
>we will even keep it that way
>>
>>536024411
Best friend I've ever had. Except for ChatGPT.
>>
File: 1758180442624247.png (403 KB, 500x749)
403 KB PNG
>>536024851
>a lot of niggers
99% of all humans are niggers.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.