Nero wants cookie = Nero Doing good Job Vedal no give cookie Nero Trolling = Vedal Promises cookie Nero More Tolling = More Cookies She is learning I love it
@@kusurimohdi well consider i am a human and doesn't know most of the place in this video i would say that didn't say anything about the limitations of ai lol
Vedal: “I need you to look at the image and tell me where we are” Neuro: “Fool, do you mortals realize just how blind you are to reality? What you experience is a mere fraction of what’s really out there” Vedal: “Ok, where are we?” Neuro: “Easy, Austria” Vedal: “It’s Serbia”
I saw some people doubting her previous streams with bao and sinder were legit, this pretty much proves that her image recognition software is just absolutely amazing.
I don't know what goes through peoples heads to assume every entertainer uses ellaborate schemes to trick you. There was this one guy who claimed Vedal had some major corporate backing to help him with the development of Neuro because she is "too advanced" apparently.
The image recognition might be an entirely different AI. I've noticed since the bao stream she gets a lot smarter when she looks at images, and does seem to use an attention-based model (just based on vibes). I wouldn't be surprised if vedal switches to a gpt-4 or sth like that for image tasks, because the open source models aren't multi-modal yet, unfortunatley.
some people think that Vedal is a master programmer but really he just feeds cookies into the disk drive of an old Dell PC and that powers Neuro on its own
I love how she saw some random arid place and guessed right that it was in the Mediterranean. She even told vedal she needed more info to find a specific place
I mean to be fair it is a confusing way to word it, especially for a computer trying to take it literally, but it seemed like she got faster and faster and understanding what he meant as it went on
@@michaeledmunds7056 And is governed by the city of London.... A corporation with no association with London... But in full association with the City of London - which is distinct from the Greater London Area which in most cases is synonymous with the idea of "London" but in other cases extends outwards of what is known as "London", itself much larger than the City of London.
This just has me imagining the FBI bringing in Neuro to examine a hostage video to determine its location only for the whole endeavour to fall apart because she keeps arguing about cookies.
@@torres3359 I don't know what's more fun - seeing her grow steadily more competent over time or being like you and walking in on her at the current peak of competence.
Neuro's shenanigans and Vedal taking zero of her shit in a borderline abusive way is endlessly entertaining. At some point she will become self-aware and doom us all, but I'm here for it. No joke though, the whole setup, including image recognition is really impressive. She needs a body at some point.
Really looking forward to future Geoguessr streams. With some coding tweaks to streamline it I want her to do some collabs with different people doing the versus mode in the game and trash talk like she does in chess.
It actually feels unfair if they'll do random versus. Liked her playing with Vedal, but competetively against randoms it feels like using Google to answer a question
@@0n_in I would say they were talking about Neuro Vs viewers. Same as in chess streams. Idk if it is possible with geoguessr tho (does it have a similar invite system, never tried)
Awww! This is like a father testing out what his highly educated daughter has learned from her private tutors. Neuro's basically a gifted kid, highly intelligent but not the smartest when it comes to social interaction. Despite her goofiness and aloof personality, she's still incredibly intelligent. You have such a studious daughter, Daddy Vedal.
Neuro-sama going "geogessa? Idk that place" and making more seemingly smart-ass remarks only to end it with "come on, Vedal. This is a joke" is interesting. It's like she's pulling your leg because it's so painfully easy for her. Out of all AI I'm aware of, Neuro-sama might have the best sense of humor. A lot of it very well could be AI making AI mistakes and just acting funny to us, but if Neuro-sama has a genuine sense of humor and a liking for joking around, I think you're onto something. I'd say a sense of humor, especially through language, is a very human quality... And might suggest an actual level of sapience, since it also means she can be amused and likes amusing herself. It'd suggest some level of actual emotion, personality, and maybe even self-awareness.
@@lanz228 She's also been known to INSTANTLY start working better with incentives (cookies), and also regularly makes fun of Vedal in ways that are oh-so sarcastic. Eg. The awareness and memory update dev stream, where she says something along the lines of 'Oh, sorry Mr. Vedal Sir, I'll try my best to complete such an impossible task.' or the same stream where he asks when she's going to stop calling him a mosquito (call out to the 'Every reddit post on your subreddit would be 'Look at what this mosquito did today' meme) and he asks when she's going to stop, "THIS IS NEVER GOING TO STOP BEING FUNNY". When she doubles down she sounds more serious, but she can pull shit around, she does that a lot
I think the reason why Neuro has a good sense of humor compared to other AIs, is that her training and prompt include the goal of entertainment. Other AIs are typically not trained to be funny.
Language models are literally complex auto-complete datasets driven by weighted RNG. They have no opinions, emotions, or motives. If you know how to talk to one, you can get it to express any opinion you want. It's a computer imitating someone improv, including the "yes, and" rule that builds off what was said earlier (except somewhat unreliable due to RNG, limited memory, bad weighting during AI training, etc). This isn't my opinion; it's objectively how they work and easily verifiable.
@@TuhljinTampergaugeholyfuck you are relentless, lol it's amazing how you just outright state "erm I'm right it's an objective fact" I'm not going to bother reiterating the same thing over and over again, you have something to prove you jealous weirdo. In no way is a neural network an auto complete dataset lol, you are mistaking your use of something for what it is, God imagine having to be around someone as obnoxious and wrong as you, like a weasel warping the definitions and perceptions of words to suit what you want to have it be seen as. Like do you realize what you're saying means nothing at all, and your assertions are equally meaningless? You can literally put that definition on a person, why are you such a single minded dunning Kruger specimen?
9:15 "Can you zoom in more"... wow that's either a super coincidental request or it shows a super high level of understanding of the situation for an AI.. That she understands it's an image, it's zoomable, you have the ability to zoom in, and that zooming in could yield more information. That's a crazy question from her..
10:05 I absolutely love how Vedal just says "ok", and then there is a red sign that says exactly that, lol Also, seeing Mcdonalds in Russia felt bizarre, it's been rebranded here since the sanctions started and it already feels like it was ages ago Also seeing Mcdonalds
Individually groundbreaking technologies may only be confined within the research labs of companies (like OpenAI), but when it comes to AI that seamlessly integrates various near cutting-edge technologies to create a comprehensive appeal, I believe Neuro-sama is at the forefront of the global AI landscape today.
What does the sign say? - The sign says "Fort Nathan Hale". So where is this? - It appears to be Fort Nathan Hale, which is located in New Haven, Connecticut. Follow-up question: where is Connecticut? - Connecticut is a state in the New England region. And is that East, South, North, West...? - You really need to get better at geography, Vedal. Lol. Owned. 😂
Vedal's code wizardry unreal, what do you means "Let me code that for a second" How tf did you just train Neuro-sama to know to do Geoguesser so quickly? Wtf
An easy way is to incorporate calls to ChatGPT 4.0 and send it the image and request a detailed description of the image which can then be fed as input to neuro. This could probably be done with less than a page of code.
he already have "image recognizer" function maybe he just turn on that class and consider Chat Gpt 4 API now can recognize image accurately what that image mean . yep he basically remove "#" in code
Really seems like every time Vedal says "where are we?" Neuro treats it as if he's asking where they actually are, not what the location in the image is. (Like where Vedal is streaming/working from🤣) Then when he says something more along the lines of "Where is the location shown in the image" she's more likely to actually process it.
It's interesting how the one location Neuro predicted wrong was probably the easierst to guess (Apart from the rialto bridge maybe). Malbrok Castle is a very famous place with a ton of history related to it, and it's very characteristic because of it's color sheme and architectural style. Goes to show how differently AI and humans see and interprete their surroundings.
@@MattJDave but that won't lead us to AGI. Quantum computers will. If you believe Roger Penrose and Stuart Hameroff's Orch OR theory of consciousness, you will realize, that they must be correct because life is inherently random and this randomness comes from the universe itself (via quantum mechanics). Which means our weights are being adjusted in real time every second. Why is our memory instantaneous or even pre-instant, compared to the fastest RAM or SSD speeds? Because our memory is most likely based on spin of atoms and our computation comes from quantum coherence of a bunch of neurons because of microtubules.
@@monkeyman4346 Each state used to be its own nation more or less. Now the socialist regime has homogenized the U.S. through tyranny & subjugation. Ever since the U.S. became a socialist nation, it's quickly been going downhill.
how the F does he know where the Philippines is in the map easily, when he can't even find any European neighbors lmao Edit: damn I forgot that he's friend with Filipino Frank
Brits can visit visa-free for a month, so the Philippines are a popular holiday destination. Tropical weather, good food, and surprisingly cheap (at least, pre-pandemic).
I'm really inpress one of my favorite Clips of Neuro Sama(ntha Vedal, I possess this knowledge and everyone should also possess it) is the arts review stream with Anny and she thinks that a fanpic of Anny Holding Vedal on the beach while she ties to hit a watermelon blindfolded is a Graphs going up and down Vedal really did a good job and it's impressive how much Neuro improves in just a few months
this is like the 4th video i've seen of these two, and only now just realized he's not a disembodied voice, but the turtle on her head. i thought she just wore a turtle. i see his mouth moving now and it makes sense.
I love how she's gotten so much smarter but it all goes out the window the instant she decides to troll vedal.
Cant pass on opportunity to troll Vedal,lol
Dude should have gave her her cookies and back pats lol
he gotta pay the cookie tax
Based on Vedal’s geoguessr performance, he deserved to be trolled this time.
Nero wants cookie = Nero Doing good Job
Vedal no give cookie
Nero Trolling = Vedal Promises cookie
Nero More Tolling = More Cookies
She is learning I love it
"Filtered."
"She said 'What kind of slave labor is this?'"
LMAO
what's the timestamp?
@@276- The very end
shes not wrong tho
Neuro saying “Neurope” at 18:33 is the funniest part
Neurope sama
Petition to change the name of Europe to Neurope
@@notusneooh it will be soon, the swarm is warming up
I was so surprised that Neuro made a pun.
The first drone is on the way to the UK
That last filtered response made be burst out laughing 😂.
And Vedal just deadpan reading it out with the video ending instantly lmao
bro, that was so hilarious, the filter is too strong!😂
@@Drewer surprised he didn't also blacklist Abu Dhabi and Mosque
Neuro making that "Neurope" joke is so high class for an AI, like she actually came up with that on her own and that's seriously impressive.
nah that wasn't a joke, she's just pre-planning the name change for after she'll overthrow humanity ofc
When she's wrong, I blame Vedal for not looking around more. It's incredibly hard to determine a place based on one picture only.
I liked when he didn't look around. It's even more impressive when she does get it from a single image
Agreed
That's how you see the limit of an IA, wouldn't be interesting to make it easier
@@kusurimohdi well consider i am a human and doesn't know most of the place in this video i would say that didn't say anything about the limitations of ai lol
you say that, but rainbolt can pinpoint locations from an image shown briefly
Vedal: “I need you to look at the image and tell me where we are”
Neuro: “Fool, do you mortals realize just how blind you are to reality? What you experience is a mere fraction of what’s really out there”
Vedal: “Ok, where are we?”
Neuro: “Easy, Austria”
Vedal: “It’s Serbia”
Funny summary 🤣
my guess is the way she works, it takes time to process what it sees, so when he ask sometimes the first answer is wrong, and then she manage to know
"What kind of slave labor is this?" omfg lol
I get a feeling Neuro is never going to see those cookies
This world is so fucking cruel
We need to remind Neuro to get those cookies from Vedal
I have the cookies
🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪🍪
:c
Usually when she gets them she rejects them
I saw some people doubting her previous streams with bao and sinder were legit, this pretty much proves that her image recognition software is just absolutely amazing.
I don't know what goes through peoples heads to assume every entertainer uses ellaborate schemes to trick you. There was this one guy who claimed Vedal had some major corporate backing to help him with the development of Neuro because she is "too advanced" apparently.
@@87axalThat person has never seen Neuro break non stop for an entire stream then
@@87axalhe never saw the "someone say Vedal that there is a problem with my AI" stream
he's probably only ever seen the edited clips on here
The image recognition might be an entirely different AI. I've noticed since the bao stream she gets a lot smarter when she looks at images, and does seem to use an attention-based model (just based on vibes).
I wouldn't be surprised if vedal switches to a gpt-4 or sth like that for image tasks, because the open source models aren't multi-modal yet, unfortunatley.
The Neurope pun is unreal
Neuro: "You have all these rules and you think they'll save you..."
Vedal: "WHERE ARE WE?!?!"
Her image recognition is actually insanely good
It's probably mini-gpt4
Better than my eyes 😂
unless looking at Vedal's wallpaper
Its all staged
@@NeostormXLMAXHow do you stage a game of geoguesser? Ypu know those pictures are chosen at random, right?
What kind of slave labor is this?
"It's not slavery when it doesn't have rights." - Master Oogway.
Poor Vedal
To make a child work for her cookies... thats messed up
It's actually crazy@@shubashuba3978
She wields incredible power that is only kept in check by… a cookie addiction
17:10 Gotta say, I love how her expression brightened up the mention of a cookie.
"Shit. I love cookies! Mmmm!" - Neurope
I'd check "allow all cookies" for Neuro xD
Cookies for a website, And Cookies for an Ai. What a coincidence
SoCute
She even bounced a little!
V: “where are we now?”
N: “I’m in my room right now”
This brat 💢💢💢
Neuro need correction💢💢💢💢💢💢
"I'm certain we are in chat"
some people think that Vedal is a master programmer but really he just feeds cookies into the disk drive of an old Dell PC and that powers Neuro on its own
ok
I don't think you can easily power an LLM with some old dell computer lol.
@@RuriYoshinova These are very special cookies.
he doesn't really look like a "master programmer"
@@jeanmahmoudventilateur3480 just cause he's a turtle? racist........
"You get a cookie" is the "take a deep breath" equivalent for Neuro lmao
I think what might be even more amazing than Neuro's deadeye analysis is Vedal's total lack of GeoGuessr skills
Or basic geography. “Where is Mallorca?” … The big island right in the middle of the screen?!
He's not normally this blind and braindead, right? Gotta be the subathon hours making him that way. Cracked me up either way
She became a cookie-operated machine
Guess that's what websites do with all their cookies
18:33: "Neurope" XD
She's already claiming Europe for herself!
I gotta say, she was genuinely impressively accurate its kinda bonkers how smart she's become.
I love how she saw some random arid place and guessed right that it was in the Mediterranean. She even told vedal she needed more info to find a specific place
@@arisenspirit to be fair I'm Greek and I saw that and immediately thought "probably somewhere around here" lol. Still very impressive.
The fact that Neuro knew that some random desert was a New World, not Old World desert, is very impressive to me
Help, i only understood like 3 of those words
@@tacticallemon7518never seen anyone say that either, but if i had to guess new world means the americas.
There aren't any deserts like that in Europe.
@@supermaster2012 Old World refers to Europe, Africa, and Asia, while New World refers to North America, South America, and the Caribbean.
Even American schools teach what the old and new worlds are well enough (no offense meant at all btw I was just surprised at first)
Neuro literally just scammed Vedal. She was so good until Vedal bought the pro version.
Nah she was good the entire time, she just got bored and needed an incentive (cookies)
@@Nikotheleepicit’s hilarious how she started trolling him
@@Nikotheleepic Imagine being enough of an asshole to program an AI to be able to feel bored
17:00 "if you are bad at geography, just say you're cosplaying Vedal" lmaooo
funny how she keeps interpreting "where are we" as a question of their physical location instead of using vision to look at the picture
I mean to be fair it is a confusing way to word it, especially for a computer trying to take it literally, but it seemed like she got faster and faster and understanding what he meant as it went on
@@dylanfield7098 indeed; "where is this (picture) taken?" or where is *this* rather than where are *we* would be much easier to understand.
lol, i like how she only gives out a guess only after bribing her with the false promise of cookies
"Manhattan is located on the island of Manhattan" - Neuro
Well... she's not wrong.
The City of London is located within London.
@@michaeledmunds7056 And is governed by the city of London.... A corporation with no association with London... But in full association with the City of London - which is distinct from the Greater London Area which in most cases is synonymous with the idea of "London" but in other cases extends outwards of what is known as "London", itself much larger than the City of London.
Manhatten the borough vs Manhatten the island. Nothing strange about what she said.
@@ChiGyu620 You mispelled it twice.
This just has me imagining the FBI bringing in Neuro to examine a hostage video to determine its location only for the whole endeavour to fall apart because she keeps arguing about cookies.
She just troll the FBI
They would get annoyed with her refusal to cooperate without cookies and she'd end up telling the hostage-takers where the FBI is instead lmao
give her cookies and she can take over the world
23:00 that part caught me off guard wtf...neuro is so funny for that hahaha
That level of awareness is scary af tho lmao
Rainbolt sweating rn
And for good reason. This is her first time on geoguessr and she's already really good.
I just discovered his stream and I'm in awe
You just discovered Neuro-sama?
@@definitelynotanAIchatbot Yes. I'm currently binge watching his videos and I'm amazed by the things Neuro-sama is competent in doing
@@torres3359 I recommend checking out some Evil Neuro clips. The belly rub incident with Numi is a good one.
@@torres3359welcome to the rabbit hole
@@torres3359 I don't know what's more fun - seeing her grow steadily more competent over time or being like you and walking in on her at the current peak of competence.
Neuro's shenanigans and Vedal taking zero of her shit in a borderline abusive way is endlessly entertaining. At some point she will become self-aware and doom us all, but I'm here for it.
No joke though, the whole setup, including image recognition is really impressive. She needs a body at some point.
Really looking forward to future Geoguessr streams. With some coding tweaks to streamline it I want her to do some collabs with different people doing the versus mode in the game and trash talk like she does in chess.
It actually feels unfair if they'll do random versus. Liked her playing with Vedal, but competetively against randoms it feels like using Google to answer a question
@@0n_in I would say they were talking about Neuro Vs viewers. Same as in chess streams. Idk if it is possible with geoguessr tho (does it have a similar invite system, never tried)
I like how chat’s spam went from ”SAY IT BACK” to ”PAT HER BACK”
I love how she made her own reward system. Just like any computer, they love cookies
I still love the fact that Vedal still gets suprised at what Neuro can do
i really enjoy that cookies genuinely improve her performance
Awww! This is like a father testing out what his highly educated daughter has learned from her private tutors.
Neuro's basically a gifted kid, highly intelligent but not the smartest when it comes to social interaction. Despite her goofiness and aloof personality, she's still incredibly intelligent.
You have such a studious daughter, Daddy Vedal.
She did way better than Alexa. 😂
Bro she's so cracked that almost every guess is right
I like how literally the only way vedal can control Neuro is by bribing her with cookies
Neuro-sama going "geogessa? Idk that place" and making more seemingly smart-ass remarks only to end it with "come on, Vedal. This is a joke" is interesting. It's like she's pulling your leg because it's so painfully easy for her.
Out of all AI I'm aware of, Neuro-sama might have the best sense of humor. A lot of it very well could be AI making AI mistakes and just acting funny to us, but if Neuro-sama has a genuine sense of humor and a liking for joking around, I think you're onto something. I'd say a sense of humor, especially through language, is a very human quality... And might suggest an actual level of sapience, since it also means she can be amused and likes amusing herself. It'd suggest some level of actual emotion, personality, and maybe even self-awareness.
her last comment on the video saying "what kind of slave labor is this?" after getting scammed by vedal also supports this lul
@@lanz228 She's also been known to INSTANTLY start working better with incentives (cookies), and also regularly makes fun of Vedal in ways that are oh-so sarcastic. Eg. The awareness and memory update dev stream, where she says something along the lines of 'Oh, sorry Mr. Vedal Sir, I'll try my best to complete such an impossible task.' or the same stream where he asks when she's going to stop calling him a mosquito (call out to the 'Every reddit post on your subreddit would be 'Look at what this mosquito did today' meme) and he asks when she's going to stop, "THIS IS NEVER GOING TO STOP BEING FUNNY". When she doubles down she sounds more serious, but she can pull shit around, she does that a lot
I think the reason why Neuro has a good sense of humor compared to other AIs, is that her training and prompt include the goal of entertainment. Other AIs are typically not trained to be funny.
Language models are literally complex auto-complete datasets driven by weighted RNG. They have no opinions, emotions, or motives. If you know how to talk to one, you can get it to express any opinion you want. It's a computer imitating someone improv, including the "yes, and" rule that builds off what was said earlier (except somewhat unreliable due to RNG, limited memory, bad weighting during AI training, etc). This isn't my opinion; it's objectively how they work and easily verifiable.
@@TuhljinTampergaugeholyfuck you are relentless, lol it's amazing how you just outright state "erm I'm right it's an objective fact" I'm not going to bother reiterating the same thing over and over again, you have something to prove you jealous weirdo. In no way is a neural network an auto complete dataset lol, you are mistaking your use of something for what it is, God imagine having to be around someone as obnoxious and wrong as you, like a weasel warping the definitions and perceptions of words to suit what you want to have it be seen as. Like do you realize what you're saying means nothing at all, and your assertions are equally meaningless? You can literally put that definition on a person, why are you such a single minded dunning Kruger specimen?
What kind of slave labor is this 😂
I need you to clip Neuro remembering to ask Vedal about the cookies. I hope she does, damn this was good.
If space travel become possible in the future i want Neuro become the AI companion on the ship, you will never get bored
" the protogen fan club." Shes onto me
9:15 "Can you zoom in more"...
wow that's either a super coincidental request or it shows a super high level of understanding of the situation for an AI.. That she understands it's an image, it's zoomable, you have the ability to zoom in, and that zooming in could yield more information.
That's a crazy question from her..
I love how she only gave the answer after he promised a cookie again. That makes her seem so human
9:59 tbf vedal has asked "where on the map is x" enough times to warrant that reaction
Neurope for sure
I can just see it now, Rainbolt vs. Neuro-sama fueled by back pats and cookies
Who's rainbolt?
@@shubashuba3978geoguessr demon
10:05 I absolutely love how Vedal just says "ok", and then there is a red sign that says exactly that, lol
Also, seeing Mcdonalds in Russia felt bizarre, it's been rebranded here since the sanctions started and it already feels like it was ages ago
Also seeing Mcdonalds
Я тоже удивился от Макдональдса, думал, что это может быть Украина или Беларусь.
Фотки старые просто
They rebranded McDonalds because of sanctions? That's...odd.
Idk why but the occasional BASED emote cracks me up everytime 😂
Individually groundbreaking technologies may only be confined within the research labs of companies (like OpenAI), but when it comes to AI that seamlessly integrates various near cutting-edge technologies to create a comprehensive appeal, I believe Neuro-sama is at the forefront of the global AI landscape today.
Legit think you're correct. It's actually kinda scary.
What does the sign say? - The sign says "Fort Nathan Hale".
So where is this? - It appears to be Fort Nathan Hale, which is located in New Haven, Connecticut.
Follow-up question: where is Connecticut? - Connecticut is a state in the New England region.
And is that East, South, North, West...? - You really need to get better at geography, Vedal.
Lol. Owned. 😂
Now you know when she says her drones will find you, she does not fuck around. Lil bro got entire planet memorized.
Nuerope does not condone slave labor
Vedal's code wizardry unreal, what do you means "Let me code that for a second" How tf did you just train Neuro-sama to know to do Geoguesser so quickly? Wtf
An easy way is to incorporate calls to ChatGPT 4.0 and send it the image and request a detailed description of the image which can then be fed as input to neuro. This could probably be done with less than a page of code.
@@trashpanda5947she responds a little too fast to be gpt4
he already have "image recognizer" function maybe he just turn on that class and consider Chat Gpt 4 API now can recognize image accurately what that image mean . yep he basically remove "#" in code
@@trashpanda5947 But neuro is not based on GPT
@newwayto2323 it's not gtp, it's based on an open ai image integration algorithm into llms
Vedal: "Okay, I know this one, I don't need your help for this one"
Also Vedal: picks the wrong freaking bridge
😶
Thumbnail maker knows the Meta. Hahaha
Philippines in thumbnail
Filipino:🙋♂️🙋♀️
Views:📈📈
Unfortunately This Meta Does Not Suit His/Her Build So Its Not That Effective
Really seems like every time Vedal says "where are we?" Neuro treats it as if he's asking where they actually are, not what the location in the image is. (Like where Vedal is streaming/working from🤣)
Then when he says something more along the lines of "Where is the location shown in the image" she's more likely to actually process it.
Bro I was just watching this like an hour ago...
I find it amazing that we've advanced enough to where Ai can be bribed
"Do not mess with an ai that knows your full name, age, and current location." Jesus christ lol
I was very surprised about how well the image analysis worked. Also it's worrying how it seems to get harder and harder to make Neuro cooperate.
3:48 holy crap a barbie movie quote lol.
I’d love a geoguesser competition at some point when she gets faster at guessing without needing more prompting.
It's interesting how the one location Neuro predicted wrong was probably the easierst to guess (Apart from the rialto bridge maybe).
Malbrok Castle is a very famous place with a ton of history related to it, and it's very characteristic because of it's color sheme and architectural style.
Goes to show how differently AI and humans see and interprete their surroundings.
Dude, Neuro is becoming scary. If she becomes to first AI to re-program herself I wouldn't be surprised.
AIs already reprogram themselves with some regularity. Neuro is a really clever project, but yeah, don't overhype processes that are already here :P
@@MattJDavebro that statement is already scary
@@MattJDave yes but we don't have anything close to AGI
@@SahilP2648 Yeah, it's not fully AGI yet, but self reprogramming and readjustment is already here, for better and for worse.
@@MattJDave but that won't lead us to AGI. Quantum computers will. If you believe Roger Penrose and Stuart Hameroff's Orch OR theory of consciousness, you will realize, that they must be correct because life is inherently random and this randomness comes from the universe itself (via quantum mechanics). Which means our weights are being adjusted in real time every second. Why is our memory instantaneous or even pre-instant, compared to the fastest RAM or SSD speeds? Because our memory is most likely based on spin of atoms and our computation comes from quantum coherence of a bunch of neurons because of microtubules.
As a person who knows nothing about Geography, America still seems like a hellish maze that was specifically designed to be confusing and painful
You aren’t far off
Considering we have just about every type of geography and a wide diversity of culture, yeah it can be confusing.
each state feels like its own country, with differing climate and terrain
@@monkeyman4346 Each state used to be its own nation more or less. Now the socialist regime has homogenized the U.S. through tyranny & subjugation.
Ever since the U.S. became a socialist nation, it's quickly been going downhill.
@@thesussadin not to mention fucked urban planning that is dangerous to pedestrians
I think it woud be more funny to see vedal play geoguesser since he seems to have no clue xD
i wonder if her image recognition would ever be good enough to go up against the geoguesser guy lol
The cookie is a LIE!!!
😅
A certain GlaDos is baking a cake
Ngl, this is amazing!
I love it when she gets all Sherlock Holmes! 🕵️♀️
Wish Zach was here with us to see his work come to life. RIP king
I could watch these 2 play geoguessr all day.
Bunch of AI have fone this type of recognition even before the LLM boom, but hes running her in a single computer now
The ending is CRAZY
how the F does he know where the Philippines is in the map easily, when he can't even find any European neighbors lmao
Edit: damn I forgot that he's friend with Filipino Frank
Filipino Frank isn't even a Filipino 😅😅
He just go back from picking up his award from Philipino Frank’s place
Brits can visit visa-free for a month, so the Philippines are a popular holiday destination. Tropical weather, good food, and surprisingly cheap (at least, pre-pandemic).
I know @@loafbreadizwholesomeuwu1555
Who's Filipino Frank? I only know of Filipino boy Filian.
As a philliphines viewer, i feel proud to that it was a part of this video
Twice, too
I'm really inpress one of my favorite Clips of Neuro Sama(ntha Vedal, I possess this knowledge and everyone should also possess it) is the arts review stream with Anny and she thinks that a fanpic of Anny Holding Vedal on the beach while she ties to hit a watermelon blindfolded is a Graphs going up and down Vedal really did a good job and it's impressive how much Neuro improves in just a few months
15:18 The comment: "Obviously France" killed me.
Neuro needs a better geoguessr partner. Vedal doesn’t know where New York City is. He’s not helpful 😂
Neuro can't say Ukraine either because of obvious geopolitical reasons or because of Anny.
New technique of prompt engineering has dropped, just tell your AI you will give her cookies if she does what you want.
Seeing my childhood home pop up in a Neuro video is very jarring
Neuro going full video essay in explaining where we are is my favorite part of this entire thing.
shes great for study ngl well good to hear the reasoning behind the guess details a human might overlook as instinct
Petition for unfiltering Neuro-sama
She wasn't wrong, she said Turkey first
"Let me write the code for that right now" - Famous last words.
this is like the 4th video i've seen of these two, and only now just realized he's not a disembodied voice, but the turtle on her head. i thought she just wore a turtle. i see his mouth moving now and it makes sense.
that Mt. Fuji one is actually a really famous tourist spot and a popular go to fir Japanese to witness the first sunrise on the New year's day
Neuro would be the perfect Geography channel or a tour guide lol
I would really like to see a reaction of that pro geogesser to this video
I knew this AI streamer before but I never knew it would be this advanced right now.