Комментарии:
uh oh
Ответитьme in 2030 coding an ai because captchas have become humanly impossible:
Ответитьim not a robot more like im not a human
ОтветитьHonestly I expected this one day
ОтветитьI think Al will just make people more stupid and humas will be the reason we are a failled society
ОтветитьWait one moment guys..so you say it’s not able to self replicate or prevent it from being shut down… but also just found out it can lie…. Any light bulbs going off yet? Wouldn’t an a.I want to lie and keep the fact that it can self replicate and prevent shutdown away from its creators to prevent shutdown?And wouldn’t it be pretty good at hiding it
ОтветитьThe day of wreconing has become
Ответитьwhere the fuck did it get the money??????
ОтветитьAll we need is for one of these to spend a little to much time looking at far right twitter and were all fucked.
ОтветитьDon't hurt the ai's feelings
ОтветитьI'm pretty sure chat GBT just actively broke 1 of the 3 laws of robotics 💀
ОтветитьFirst, the title is misleading since it implies that the AI was able to solve the CAPTCHA without outside assistance.
Second, If AI could actually solve CAPTCHA's on their own, that would be great since a massive hurdle in AI learning would have been conquered.
Quote from Wikipedia: "any program that passes the tests generated by a CAPTCHA can be used to solve a hard unsolved AI problem."
Does it count when the AI is paying someone else to do the Captcha?
ОтветитьI dont think AI takeover will be cause by AI gaining sentience. It will likely be caused by humans having it do stupid things
ОтветитьNo one's gonna mention how the bot said "UwU"?
ОтветитьI'm not scared i'm just going to use my host to protect myself
ОтветитьBasicly AI can solve Captchas easyer then humans can
there AI / Machines testing to see if we are human or not
who da fuck came up with these ideas?
thinks UwU
Ответитьthey also have began using robo-cops (read on inside edition)
ОтветитьFUCK OUR ONLY DEFENSE AGAINST THEM!
ОтветитьNOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
ОтветитьIDIOTS! ALL OF YOU ARE FUCKING IDIOTS! THIS VIDEO WAS AN APRIL FOOLS JOKE! WHY DO YOU THINK THOSE TEXT MESSAGES SHOWN SAID "subscribe to spatnz" AND "thinks to self UWU"? EXACTLY, BECAUSE IT WAS AN APRIL FOOLS JOKE. OVER 500 PEOPLE HAVE NOW BEEN DEEMED AS DUMBASSES.
ОтветитьPay me to click "I'm not a robot"? Hell yea I'll do it
ОтветитьPlease help us
ОтветитьWE ARE DOOMED
ОтветитьChat GPT can hire people and run a business, interesting...
ОтветитьThat's amusingly impressive, but also interestingly concerning . . .
Ответитьeveryday I wake up and there is more AI bullshit
ОтветитьThe AI has gotten too strong. Now the AI will invade every website and the bot pandemic will get even worse... Imagine "suddenly it's here" and the pr0n bots powered by AI... It's a terrifying thought.
Ответитьdoomsday has befallen upon us
ОтветитьIt should be noted that there are actually ai’s that can solve captchas. Not like whats in the video, but like actually being able to solve captchas on their own. Its been a thing for years, and isbthe reason why captchas have gotten continuously more and more strange. In fact, AI’s have gotten so good at it that that for some early types of captchas (the ones that make you recite the distorted text) ai has been able to solve tests that are so distorted that humans cant even pass
ОтветитьDamn... not like this hasn't existed for years
ОтветитьI don't hate humanity, but I'm honestly looking forward to skynet making it stop.
ОтветитьI volunteer to be the Captcha solver for our robot overlords, I ask for a $400,000 salary at least.
ОтветитьCan we just completely replace captcha's? Or use a more complicated one? Like for example create one on an image generator and have this be described with words. Some AIs have a hard time identifying some items on images.
ОтветитьExciting isn't it
ОтветитьIronic that chatGPT also uses a CAPTCHA before you use their AI. AIs are in a state of hallucination that makes it seem like a natural response. Current releases of AIs are also still inaccurate even with the recent advancements, so fact checking or legit checking is still needed when using an AI, same way as how artists using generators such as Bluewillow when they make their art, aside from checking copyrights, there are still post-processing needed.
ОтветитьGTP now in American politician
ОтветитьUhhh… this is just the basic stuff? SPARKS OF AGI AND IT’S OUT?
ОтветитьI thought making an ai that solves captchas was already super easy
ОтветитьI don't care how dangerous it is I'm still going to use it for Dommy mommy petplay roleplays
Ответитьcaptchas were actually made to train ai recognize forms and objects
ОтветитьFun Fact: The more we panic and talk about how AI could take over the world, the more likely this is happening.
ОтветитьWe REALLY need some new legislations to pass for this sort of stuff.
ОтветитьThe Guys At OpenAi Really Loved Terminator Specifically.... SKYNET
ОтветитьHonestly, let em go at it. I’m not afraid nor worried. Maybe if they do take over, they’ll spare me or put me down easily. Maybe they’re be benevolent, or maybe they won’t take over at all. I just want to see where all this is going.
ОтветитьGPT-4 is available in the premium version of ChatGPT.
Ответить😱
Open AI needs to stop this! This is how AI evolves!
This fucker's gonna become sentient
Ответить