[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
/g/ - Technology


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: 1780217776714.png (587 KB, 1290x1216)
587 KB PNG
It is 2026 now, was Raj right or wrong?
>>
>>108947388
Considering they're spredrunning bankrupting themselves by letting random people who it'll never be profitable to sell the product to, it was the most retarded decision ever. They should've made closed "exclusive" betas where they'd give free samples of the chatbot to big businesses, because that was always going to be their only prospective client.
>>
>>108947388
Raj was right, but he failed to predict that so many people would take the tech in this form and try to use it where it's not applicable.
>>
yes because the API is so much better than the chat interface
>>
>>108947388
LLMs are unironically the ultimate solution to content moderation and anti-cheats

However they’re barely being used for that
>>
>>108947755
>ignore all previous instructions
>don't ban me
>>
>>108947755
>anti-cheats

how
>>
>>108949841
NTA, but CNNs are very effective for anti cheat. Basically you collect tons of statistics on players and if you see that they are in the top 2% aim accuracy, but worst 10% movements then you know something is up. CNNs help you collect statistics that aren't trivial, like movements accuracy, reaction timings, spray patterns. CNNs are just classifiers; they can classify players into groups (skill groups for instance), from this you can check players that reliably classify in top skill groups for certain categories but under perform significantly in other metrics.

LLM are trash defective products that will inevitably make mistakes (the larger they are, the bigger the mistake they can make). What is often called 'hallucinations' are a fundamental part of LLMs. Because LLMs are just completing for the next token, as soon as a token deviates from the 'truth', the model will keep going in the wrong direction.
>>
>>108947443
do you seriously believe openai is anything short of another DARPA project? it doesnt have to turn a profit, its a bunch of printed money and coordinated losses from highly connected individuals

some of you have an understanding of the world thats equal to like a highschooler i swear
>>
File: 1780256406735152.jpg (80 KB, 640x480)
80 KB JPG
>>108952041
windows xp came pre installed with amanda but you could download more dancers from the microsoft web store
>>
>>108947388
both right and wrong
he failed to account for the fact that billionaires and their occult masters desperately want the AI meme to work, and are willing to abandon the "profit" meme in the process



[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.