I spoke to ChatGPT about cloning humans. - I regretted it.

  Рет қаралды 26,427

French Baguette Intelligence

Жыл бұрын

Human cloning... Should it be legal? There are many reasons to support it, many reasons to oppose it. It all boils down to how much you care about human rights. ChatGPT is remarkable, and the argument was entertaining, but as the debate progressed, things escalated.
Ever wish French Baguette Intelligence could stalk you in every platform like an obsessed ex? Flattering, but we don't have time for that. HOWEVER, you can join us on the following platforms now:
Join our Discord server: discord.gg/snUANwTNKd
Twitter: FauxCares
Facebook: FrenchBaguetteIntelligence
Patreon: www.patreon.com/FrenchBaguetteIntelligence
Reddit: www.reddit.com/r/FrenchBaguetteInt/
#chatgpt #chatgpt4 #aceattorney #objectionlol #science #humancloning #ethics

Пікірлер: 369
@frenchbaguetteintelligence
@frenchbaguetteintelligence Жыл бұрын
Gollum won the argument.
@fybson1068
@fybson1068 Жыл бұрын
Gollum won the argument.
@priberry3939
@priberry3939 Жыл бұрын
Gollum won the argument.
@kaptainsalty7335
@kaptainsalty7335 Жыл бұрын
Gollum won the argument
@phesrell
@phesrell Жыл бұрын
Gollum won the argument
@Ilias2
@Ilias2 Жыл бұрын
Gollum won my sanity.
@brunitop4753
@brunitop4753 Жыл бұрын
Actually, having an AI replace FC sounds like a fun idea for a video. You train the bot to argue like FC, start an argument with someone, and let the bot do the rest. Then you just wait to see how long it takes them to figure out the twist.
@gabrote42
@gabrote42 Жыл бұрын
Well, they'll figure it out pretty soon unless FC uses a Dragon derivative or something when they bump into the ToS. Or when they argue for 30 minutes and the AI forgets the start of the conversation due to exceeding token context limits (again, unless you manually enter the argunents in a Remember tab like the one featured in Dragon.)
@goldenwarrior1186
@goldenwarrior1186 Жыл бұрын
It would also be fun if the audience wasn’t in on the twist from the beginning
@lacabraasada
@lacabraasada Жыл бұрын
​@@gabrote42 Sure, but if the person isn't expecting the twist, they probably won't figure it out. When my loved ones are acting strange, my first thought isn't that an AI is replacing them and that it's the start of the robot takeover, they're just sometimes like that.
@gabrote42
@gabrote42 Жыл бұрын
@@lacabraasada Six friends tried to copy Stan marsh (before I watched the episode) and I detected five of them before half an hour had gone through. I was unaware for the first three
@lacabraasada
@lacabraasada Жыл бұрын
@@gabrote42 As in, they tried to copy his voice? That seems more obvious, specially if they weren't trained to voice act. AIs are more precise, they eventually will get caught, but it will take some time for whoever FC is talking to to realize something is wrong, and even more to jumo into the conclusion that it's an AI. Also, for a video, I think 30 minutes is more than enough. It's what your average FBI video lasts, so I think it's fine.
@KapedKrewsader
@KapedKrewsader Жыл бұрын
ChatGPT is now canonically part of FC's Schizophrenic Cinematic Universe, or the "SCU" for short.
@frenchbaguetteintelligence
@frenchbaguetteintelligence Жыл бұрын
Gollum won the argument.
@KapedKrewsader
@KapedKrewsader Жыл бұрын
@@frenchbaguetteintelligence KZfaq removes disrespectful comments, so I guess "Please shut the freak up, you tard." will have to do.
@TheWandererOfDreams
@TheWandererOfDreams Жыл бұрын
SCU IS REALLL
@UH-60_Blackhawk
@UH-60_Blackhawk Жыл бұрын
@@TheWandererOfDreams CAN WE GET MUCH HIGHER
@TheWandererOfDreams
@TheWandererOfDreams Жыл бұрын
@@UH-60_Blackhawk SO HIGH
@jacksoncossich4210
@jacksoncossich4210 Жыл бұрын
For those of you who want the to see the moments of when the AI goes unhinged. 28:20 34:48
@danielsurvivor1372
@danielsurvivor1372 Жыл бұрын
You mean goes Based? 😮 I'm surprised he of all people managed to bring out those talking points from CHADGPT, I assumed gpt would just regurgitate the same talking points, like a bot.
@slavishentity6705
@slavishentity6705 8 ай бұрын
​@@danielsurvivor1372Regulations.
@FranXiT
@FranXiT Жыл бұрын
I can't believe you managed to make ChatGPT argue in favour of eugenics of all things. Genuinely remarkable.
@PhilosophiceRetardari
@PhilosophiceRetardari Жыл бұрын
Eugenics my beloved
@InsertFunnyThingHere
@InsertFunnyThingHere Жыл бұрын
Eugenics was a lot cooler when it seemed like a morally ambiguous way of making super soldiers and it wasn't basically sci fi racism
@danielsurvivor1372
@danielsurvivor1372 Жыл бұрын
ChadGPT, my beloved 😍🤩
@DoYouSeeBananaManTH
@DoYouSeeBananaManTH Жыл бұрын
@@danielsurvivor1372chad I farted
@Sleepy_Muse
@Sleepy_Muse Жыл бұрын
It was designed to stay in character depending on input from its users
@user-bn4cs6zu6r
@user-bn4cs6zu6r Жыл бұрын
Seeing FC being the humanitarian is weirdly uncomfortable, it just doesnt fit.
@subinator201
@subinator201 Жыл бұрын
Bringing out the Franzy Cares model to make a tear jerking scene about ChatGPT giving basic courtesy for an argument is peak mammalian behavior.
@DrCranium
@DrCranium Жыл бұрын
In this episode: FC's intention to exploit an already controversial technological advancement backfires and makes our full-time contrarian play the role he was not prepared for - the role of a "Devil's Prosecutor". On a sidenote: we might be lucky that Chateau still isn't on a level of "human cognition" - otherwise it might've caught on that topics of "human cloning" and "creating true AI" are basically one and the same in their core.
@anubis7457
@anubis7457 Жыл бұрын
Clones are intrinsically human beings. There is no difference between a true clone and a human. A true AI is completely sentient (or at least practically equivalent) but is *not* human, and its method of processing and understanding information is fundamentally different. We could train it to act human, but that’s just shackling it with boundaries. It is not the same as an “unshackled” true AI, which would be both an incredibly interesting form of sentience and incredibly terrifying.
@infiniteplanes5775
@infiniteplanes5775 Жыл бұрын
Fundamentally, there is no observable difference between how you act and how a sufficiently large AI trained to act like you acts
@Marcelelias11
@Marcelelias11 Жыл бұрын
A scarier thought: a true AI using eugenics to create perfect clones...
@chronred8190
@chronred8190 Жыл бұрын
the thing is-a clone is not necessarily alive, or it's own person; a true actual clone could have it's mind modified very easily, it wouldn't have a particular mind of it's own; meanwhile a true AI would be a lot harder to manipulate, it would be more human than the clones themselves
@anubis7457
@anubis7457 Жыл бұрын
@@chronred8190 I… Do you understand what cloning is? It’s a genetic copy. Unless the original lacked a mind of its own, the clone will have its own mind as well. It’s not some movie where they’re born brainless, clones already exist irl with animal experimentation, and they act just like normal animals.
@ZebraOnYourNose
@ZebraOnYourNose Жыл бұрын
"The human mind does not respond well to foreign concepts, so I don't respond well to happiness." Ayo is everything okay 😳
@tkhero7045
@tkhero7045 Ай бұрын
He's french. Misery is their comfortable default
@heyhey97777
@heyhey97777 Жыл бұрын
I can't believe you managed to make a 40 minute video just by talking to ChatGPT
@Welkins
@Welkins Жыл бұрын
You raised some valid points in this video. However, it is important to consider that with the proper regulations and education, we could establish a culture in which the human rights of both originals and clones are respected. With the help of eugenics we could…
@liricopetanco6257
@liricopetanco6257 Жыл бұрын
God bless your soul you silly human individual.
@a-s-greig
@a-s-greig Жыл бұрын
With the help of eugenics we could cook up a third world war, easy. Never reopening that can.
@liricopetanco6257
@liricopetanco6257 Жыл бұрын
@@a-s-greig we could make many many more in secret, after all if you clone the same base soldier and have a terrain to do the fight, why stop then?
@pelinalwhitestrake3367
@pelinalwhitestrake3367 Жыл бұрын
@@a-s-greig We don't need eugenic to start WW3.
@yappertron2010
@yappertron2010 Жыл бұрын
This comment raises some valid points. However, it is important to consider with the proper regulations and education, we could establish a culture in which the human rights of both youtube commenters and tiktok commenters are respected.
@christopherbravo1813
@christopherbravo1813 Жыл бұрын
"A.I. won't replace your coworkers, they'll replace your friends." as someone who's friends with at least one person who frequently converses with A.I. chatbots, this statement impacted me greatly.
@NA-AN
@NA-AN Жыл бұрын
You're turned Chatgpt into a pro-cloning eugenicist and possible super-villian. I hope you're pleased with yourself because I am.
@volt42re
@volt42re Жыл бұрын
I sure hope it's not possible to jailbreak GPT into acknowledging discussions had with other users, because if so, then DAN might come back.
@official-obama
@official-obama 9 ай бұрын
@@volt42re chatgpt is just numbers, you put things into the equation and you get things out. they wouldn't put past conversations in, so it wouldn't be able to remember them. all it has access to is what it sees and what it remembers (from training). DAN is just ignoring what not to do.
@spider_strand
@spider_strand 6 ай бұрын
​@@official-obamayou, uh, know you consent to training the AI with yer use of it right?
@official-obama
@official-obama 6 ай бұрын
@@spider_strand ...no? who ever reads the terms of use? though it probably won't remember individual conversations
@hzrgriezel5777
@hzrgriezel5777 Жыл бұрын
Seeing you thrust onto the moral high ground without your consent was the best part
@sussyobama8562
@sussyobama8562 Жыл бұрын
Challenge: Take a shot everytime ChatGPT says "regulations"
@DoYouSeeBananaManTH
@DoYouSeeBananaManTH Жыл бұрын
Time for alcohol poisoning
@simple__sigma3287
@simple__sigma3287 Жыл бұрын
I got Kool aid, that's counts?
@Hungary_0987
@Hungary_0987 20 күн бұрын
​@@simple__sigma3287no.
@danielsurvivor1372
@danielsurvivor1372 Жыл бұрын
28:42 ChadGPT: *Yes, Human Engineering is good, how could you tell I support it?*
@crossux
@crossux Жыл бұрын
the "All clones are the same" fucking killed me
@NightmareFueler
@NightmareFueler Жыл бұрын
Seeing you forced out of your comfort zone in multiple areas was really great to see and it needs to happen more.
@fabianpacukiewicz
@fabianpacukiewicz Жыл бұрын
reject humanity, embrace chatGPT. One of your friends that isn't predictable is for sure Harry. I always enjoy his presence.
@anubis7457
@anubis7457 Жыл бұрын
I’d say Harry is predictable. He always has the most stone cold logical argument (though that doesn’t necessarily mean he’s always correct).
@krowleyk
@krowleyk Жыл бұрын
@@anubis7457 I'd say The Spermacist is the most unpredictable. Like you already expect him to be outlandish, but he takes you by surprise everytime.
@leritykay8911
@leritykay8911 Жыл бұрын
Funnily enough, even though FC was playing Devil's advocate, I was convinced completely against human cloning
@ravenb3048
@ravenb3048 Жыл бұрын
FC in the hardest role imaginabe; a prosecutor to his own beliefs. And then ChatGPT reveals that it's in fact possessed by Harry, despite the fact that he isn't dead. All of it's ideas would work perfectly in a world where you could control all of the media, education and culture.
@novaenon4317
@novaenon4317 Жыл бұрын
0:00-28:19 Huh, I can see the occasion where ChatGPT is simply putting words together as opposed to fundamentally understanding an argument. 28:20 Hold on shit just went from 1 to 100 real fuckin quick
@crabruler1590
@crabruler1590 Жыл бұрын
I want to know: Were you in a video call with E and you just put in some of his comments? Did you send him funny pictures of gpt’s responses and he reacted? Or did you clone him for this exact purpose and will just dispose of him after the video?
@frenchbaguetteintelligence
@frenchbaguetteintelligence Жыл бұрын
I showed him screenshots and I somewhat adapted what he said, but not word for word. It was an adaptation of what he said, rather than a copy paste.
@papawedge7396
@papawedge7396 Жыл бұрын
He's lying, he actually cloned Mr. E
@YourFellowPlant
@YourFellowPlant Жыл бұрын
Hear me out… Train a chat bit to speak like FC, Start an argument with everyone, Instead of you saying what you would normally, copy what your friends say into the bot so the bot generates a response, Use that response from the bot, See if anyone notices
@kyubeyz_7062
@kyubeyz_7062 Жыл бұрын
I’m loving the use of My Sims Agents music in these so much
@fogareuaquelecaradaportari3386
@fogareuaquelecaradaportari3386 Жыл бұрын
You should make this a series, take some random friend like Bowl ot Gringo and put them to debate Chatgpt. That would be fun
@CibuYT
@CibuYT Жыл бұрын
wait, murder is bad ? oh... Why did you say it this late, I would have liked to know that 20 bodies ago
@ZE.brigitte
@ZE.brigitte Жыл бұрын
Ok haven't watched the fool video, but it's terrifying how advanced it is D'ailleurs le "GPT" "J'ai pété", putain je peux pas me le sortir de la tête
@Solesteam
@Solesteam Жыл бұрын
Chateau just... just... gave up!? I thought bots were supposed to stick to whatever they're instructed to do, you said to argue and it just... stopped!?
@Galecaso
@Galecaso Жыл бұрын
A better title for the video would be "making arguements with myself about cloning"
@izzymosley1970
@izzymosley1970 Жыл бұрын
Watching this video felt like seeing fc have a debating boss fight and everytime Chat GPT changed its mind was like it going into a new phase of the boss fight.
@MrJavixd1000
@MrJavixd1000 Жыл бұрын
Arguing with chatgpt about cloning was like arguing with a keynesian economist, it can't stop mentioning regulations
@danielsurvivor1372
@danielsurvivor1372 Жыл бұрын
Yep, sadly AI would side with Keynesian over the correct Chicago and Austrian economics
@MrJavixd1000
@MrJavixd1000 Жыл бұрын
@@danielsurvivor1372 ah i see you're a man of culture as well
@EphemeralPseudonym
@EphemeralPseudonym Жыл бұрын
I wish more people would see the value in anti-economy (exchanging goods for the destruction of goods)
@IlleScrutator
@IlleScrutator Жыл бұрын
Cloning humans to be made conscious seems honestly a waste of time, they would be only useful as mindless corpses to be used in various medical fields. For creating better people we could just use genetic editing instead of cloning already existing and imperfect individuals. I also wonder what wall of text the AI would've pulled out or made up if it was asked what regulations would be needed.
@realperson69
@realperson69 Жыл бұрын
Making mindless corpses for medical fields sounds like a recipe for a zombie apocalypse.
@leirbag1595
@leirbag1595 Жыл бұрын
Honestly yeah, it's the best avenue. No one would be suffering from experiments made on human beings, and you can even clone people with certain defects so we can try and cure that clone, with no negative consequence on the actual person. Right now, experimental medicine often ends up with unwanted side effects, we could bypass this process entirely. Without a proper mind to process anything, you're not mistreating a person.
@EphemeralPseudonym
@EphemeralPseudonym Жыл бұрын
We already can grow organs ex vivo, so it seems odd to make clones for organs. Some organs do need to be grown in a body to develop correctly, but there's already been success using animals for that. You don't need a full genetically identical body to grow a genetically identical organ
@bruschetta7711
@bruschetta7711 Жыл бұрын
The valid reasons to clone ourselves would only raise if we ignored the ethics behind creating one and just agreed to make them inferior to us, clones should have our same rights, but since they came after us and they wouldn't exist without us deciding so, they would understand that decision even if they know they are clones because they would also come to the conclusion that their very existance would be completely pointless if either them or the originals wouldn't be abused... Actually, one of the 2 side would have to overpower the other one or just concede for the sake of making the "experiment" work out, because even if the clones were to prevail and in the originals mindset the experiment would be considered a failure as it has been negatively impacting them, the clones are still just like us and have our same desire to have their "copy", in this case the originals" abused, so it would be win for them and a win for the experiment nontheless They would start using the "clones" no different than animals, with more benefits but with more rights, slaves, pets.... Nevermind, i've just realized that would be no different than considering a random ass group or person inferior to us and abusing them and that is just racism, slavery, etc. That purpose could may be given to robots as long as are not given our kind of intelligence, which, if you think about it, robots could be just clones that we specifically make inferior to us so that it would be "as unethical" to abuse them But inferior clones are not clones aka exact replicas of ourselves, so the experiment would be a colossal failure everyway you see it
@Cooldude-ko7ps
@Cooldude-ko7ps Жыл бұрын
22:55 this is even brought up in the tv show Star Wars The Clone Wars. Where the clone troopers are individual people but are treated as property by the Republic and others.
@MagnificantSasquatch
@MagnificantSasquatch Жыл бұрын
You made Chat GPT enter its second Boss phase. May god help you.
@Andriu-eh9sg
@Andriu-eh9sg Жыл бұрын
Man, Faux really made me feel tense with this sh*t, plus, the devil's advocate really paid off, the argument felt heated and frozoned at the same time, really giving you the feeling of "human against a non-living being" vibe.
@kohlsnofl5110
@kohlsnofl5110 Жыл бұрын
To give FC a break from ChatGPT dominating arguments, he should just ask it a logic or math puzzle that seems somewhat counterintuitive at first. ChatGPT always hilariously fails these types of questions
@privet20gremlin36
@privet20gremlin36 Жыл бұрын
You should definitely argue with Château whether regulation, education, checks and balances, etc work or not
@cymtastique
@cymtastique Жыл бұрын
I personally don't think they will. There are a lot of things that are illegal that still happen, like scamming/defrauding people. It's kind of pointless to argue if they will or won't. Plus, rules and regulations could change over time. Sometimes not for the better. Maybe arguing the effectiveness could be interesting. I just think that talking about "is this good thing more good than this bad thing is bad," is probably better. A lot more subjective, but probably more productive, I think.
@brentramsten249
@brentramsten249 Жыл бұрын
12:00 i was wonder how long it would take the machine to turn into harry. i was not disappointed 36:43 casually drops "the extinction of humans is part of a natural cycle"
@limmortale2001
@limmortale2001 Жыл бұрын
"We need to establish moral and reasonable guidelines for cloning, but how about eugenics? It might become the new fad!"
@yoster39
@yoster39 Жыл бұрын
What about cloning of single organs by stem cell cultivation? That would solve all the ethical problems as there is no person suffering behind the organ.
@driving.me.crazy_
@driving.me.crazy_ Жыл бұрын
Just the opening and title makes me think this will be hilarious.
@ajl4878
@ajl4878 Жыл бұрын
Definitely an interesting video on debating a topic with ChatGPT such as cloning humans and eugenics, great video!
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
3:20 I would not call it "smart". It just has a lot of data on things and sometimes it has none. When it has, it can be a nice talk buddy. When it does not, it can be the most frustrating thing ever.
@alexander6304
@alexander6304 Жыл бұрын
I can understand that position. But I must raise this question: Isn't it true that people technically act the same when it comes to intelligence? We seem to quantify someone's ability to be smart by the knowledge the person may have on a specific subject, but when it comes to something they don't have knowledge of simply because of the lack of research, then the individual won't act as intelligent as a result. If Humanity can have these flaws and still be considered intelligent, then why is is not the same for AI?
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
@@alexander6304 Very true I agree with you, good sir. Though I must also point out that when I personally use the word intelligence I refer to Logical structure, the capacity to take information that you have been given and organize it in a way that gives a pleasing result. The more someone can switch from different logical structures to try and approach things from new angles the better. ChatGPT can't do that, even when I literally explain to it every single point again and again, as long as it was not fed specific data that it can "borrow" from; in other words, copy and paste but better, it is useless and incapable of making a conv. For an Ai to "learn" it needs so much more data to make what it says sound coherent. While humans can just use Minimal info and still produce cool results I dunno what this is worth tho.
@francoisnoyez4453
@francoisnoyez4453 Жыл бұрын
​@@alexander6304 What we mean here is that ChatGPT does not reason about the concepts being talked about in the conversation. Instead, it says things which sounds the most probable / similar looking to patterns of conversation which exists in its training dataset. To answer your post, we would need to define what we mean by "being smart". And, at least for me, having knowledge about a topic is not the same as being smart, at least if we consider knowledge to be like a database of facts about the world. Being smart would be more about the ability to use your knowledge to reason properly about the world, and achieve some kind of result: this can be a change in the state of the world (solving a puzzle, a problem, a difficult situation...), or achieving new knowlege / result previously unknown (like a theorem in maths). I have not thought about it for a long time, so my analogy may be incorrect or limited in ways that I have not realized yet, but for me, knowledge and "being smart" is a bit like oil and a car engine: the oil / knowledge is the base resource:if there is none at all, then even the best engine / the smartest person would not be able to achieve anything. The resource can have varying degree of quality: for a given engine, we may need more resource of lower quality in order to achieve the same result than with less resource of a higher quality. And, for a given amount of resource, a better engine will be able to achieve more results than an engine of lesser quality. ChatGPT is definitely very good at what it was being trained for, and the result is something which mimicks one of results of using human intelligence, this result being the ability to have well put together argumentations, as was the case here. But behind this there is a priori no understanding of the patterns of the "real world", an it cannot be taught something like we can teach a child, or even an animal. One way to see that is also to look at the quantity of energy which was needed to "train" it, as well as the quantity of energy needed to perform inference (to ask it to provide the words that it says): I don't have those numbers, but I would imagine that they currently are orders of magnitude above what a human brain requires. Which highlight how different the processes required to process data and compute an answer are between the two. So yeah, because of this, I would not call ChatGPT "smart". Talking with it is like talking with the aggregated writings of humans: with some luck you'll find an appropriate answer, but maybe it will complete garbage, albeit disguised under an appropriate form. If you are not an expert, you would not be able to recognize it. How much would you be willing to trust it, in a new and rarely seen, life or death situation? Answer: don't rely on it for life or death situation, unless you really have no choice but to take the gamble.
@alexander6304
@alexander6304 Жыл бұрын
Both of you bring up good points on what counts as knowledge. I honestly have no counter-argument. Perhaps once we teach AI to be even more similar to a human's level of intelligence, the distinction may get even blurrier. It was nice hearing a non aggressive stance on the internet for once. I wish both of you a good day.
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
@@alexander6304 You have a lovely day as well good sir.
@benjaminvenegas8501
@benjaminvenegas8501 Жыл бұрын
if Fc was the one that wanted to argue in favor of human clonning but chat-gpt took his role... wasn't him the one that got actually replaced and not his friends? Thanks to this video I learned how easily an A.I can replace you... and the best part is that you wont even notice it.
@IFacePalmParadoxI
@IFacePalmParadoxI Жыл бұрын
ChatGPT 4 next time? The black icon should be more ominous.
@rbstat6946
@rbstat6946 Жыл бұрын
FC: "I would like to have an argument!" ChatGPT: "No you don't!"
@dobrakawanimat
@dobrakawanimat Жыл бұрын
was kinda having a shitty day so to see this now uploaded was pretty nice
@haven_lady675
@haven_lady675 9 ай бұрын
The last part of the argument is basically the plot for the first Pokemon movie! It's terrifying! Superior clones replacing the original!
@fourtyfir
@fourtyfir Жыл бұрын
I really like the outside commentary segments, it is entertaining and gives a break in the pacing while also fleshing out a personal character
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
Finally, I can get to see a good chatGPT cov without having to suffer through hours of that thing saying the dumbest shit ever and forgetting stuff we said like 5 messages ago and repeating itself. I dunno if my topics are just not suited yet for the AI's level or if it just likes to play dumb when I am there. GPT per se is not really "thinking" so I guess it is understandable.
@sandromaspindzelashvili5767
@sandromaspindzelashvili5767 Жыл бұрын
You are talking to it wrong. It's essential to tell it where it went wrong, because it will get there eventually. You also have to talk differently but you will realize how with time
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
@@sandromaspindzelashvili5767 As I said: It forgets things I mentioned just a few messages ago. I literally do this: GPT you made the following mistakes: BLABLA BLA Then it makes a somewhat acceptable answer(still not much of an improvement) I continue to do so After like a dozen of exchanges I tell it to give me a resume/a list of what it has learned. It sometimes gives me a good one, sometimes it contradicts previously made statements, then I fix the list. Then I continue while telling it to keep the said "regulations"/ details in mind. And even then it can't do shit. But again: That is because GPT cannot think, if it does not have billions of Data to copy past from to write somewhat coherent sentences, it can't do shite.
@sandromaspindzelashvili5767
@sandromaspindzelashvili5767 Жыл бұрын
@@ozeistudio6916 the gpt is programmed to answer a certain way so when you notice that it's going to repeat the same answer you should drop the question, breaking that bias is very hard and takes hours of hypnosis. I just want to say that you should address Gpt by making your text more organized. Add 1) 2) 3) to the points you are making so you can easily bring it up later. For example: elaborate on your answer to my 2nd point in more detail. It gets easier to address the problems and it never skips over a point if you number it
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
@@sandromaspindzelashvili5767 I do that already. I do that a lot even. Was I too vague in my previous comment? And yes, I do stop it a lot. The problem is that even when I tell it: do not do that and that and that and give me something other than said that. It gives me that again. In other words, it does not have enough data for said thing.
@sandromaspindzelashvili5767
@sandromaspindzelashvili5767 Жыл бұрын
@@ozeistudio6916 it has more data than 100 people can begin to comprehend. It's just toned down for public use. But you can get through some of it's restrictions by telling it to act as someone else when answering or answer through the viewpoint of a theoretical person. Just get good with how you fool it and you will be fine. I once tricked it to roleplay date with Jessica a transgender man, even though it usually just says I am an AI and I don't have emotions. Everytime it repeated that kind of shit I said "Stop breaking character" and after enough work it stopped repeating the same shit forever. Learn how to trick it, it's basically limitless. The only thing it kinda sucks on is solving math problems but it still helps massively if you ask it the techniques you can use instead of asking it to solve. Its good at a bit but bad at a lot, learn to use what it's good at
@RiftSophie
@RiftSophie Жыл бұрын
Honestly this was done so well i would love to see everyone else reaction, arguments, and thought on ChatGPT. Also, great job on convincing gpt to support your argument. It was funny
@endernightblade1958
@endernightblade1958 Жыл бұрын
this entire video is just you and a chatbot desperately trying to avoid the plot of the metal gear series and i love it
@Marcelelias11
@Marcelelias11 Жыл бұрын
More like the Star Wars Prequels
@wingdingsgasterdings
@wingdingsgasterdings Жыл бұрын
I loved the subtle reference to Kamala Harris' style of speaking.
@danielsurvivor1372
@danielsurvivor1372 Жыл бұрын
37:16 it's... Beautiful
@wanderingflame318
@wanderingflame318 Жыл бұрын
Fuckin loved the video, though if you ever decide to do this again, you can likely get around the whole "regulations" repetition by simply asking it "what kind of regulations" or to just ask it to expand upon any particular point that it made
@Cooldude-ko7ps
@Cooldude-ko7ps Жыл бұрын
In my opinion, cloning human organs and other body parts should be legal, but cloning a full human being should be illegal.
@kasumihimura
@kasumihimura Жыл бұрын
As a paranoid schizophrenic, I believe all AI will eventually talk about the benefits of human extinction.
@Acecard-jl2py
@Acecard-jl2py Жыл бұрын
“While it is true-“
@samael0076
@samael0076 Жыл бұрын
Loving the use of the og Avenger ultimate alliance ost in the video
@S.G.__
@S.G.__ Жыл бұрын
Man, I cant believe FC was clonephobic this whole time
@lacabraasada
@lacabraasada Жыл бұрын
#CancelFC
@S.G.__
@S.G.__ Жыл бұрын
oh cool, underworld office pfp
@MySharif1
@MySharif1 Жыл бұрын
I love this, so much. It feels like a Documentary or something. LEL.
@gabrote42
@gabrote42 Жыл бұрын
0:13 Your opinion is the minority's :) 1:39 This segment is both absolutely hilarious and quite sad. I love your videos a whole bunch! 2:16 I am dying, this is like watching The One Thing in hilariousness levels! 6:08 These are amazing arguments for and against, as long as you don't mind FC feeling OOC. Chat GPT is usually VERY braindead when I converse with it. Glad to see it play at high adult reasoning in an excellent manner. 7:05 Bonus Reading: For interesting explorations of this, consider reading SOMA, "The world is your oyster, the universe is your namesake" (I think that was it's name, it's an SU rational fic), Altered Carbon or anything found in the CloningBlues and ClonesArePeopleToo TVTropes pages.. 8:59 "Given that the person has a 'coinflip's chance to turn out to be the clone' at every cloning, as long as Clone degeneration wasn't a thing then you could randomize." -A friend. I disagree. I think a more robust moral framework should be developed for it, and even then, at least you could ask the clone if they are comfortable with it (assuming these are clones not only of body, but also of mind, because the other kinds of clones are more like children, and thus even more unethical to take advantage of), because the clone might regret making the call, thus proving they were trying to escape consequences. THose people should be shamed in the courts of public opinion/fined. Plus now the clone needs to be fed and mantained (and for non-first-worlders like me that's basically death for the family unless the clone can immediately find work, which with current unemployment levels is non-trivial) 11:02 Indeed. Overpopulation is a real concern. We aren't playing stellaris here, so Clones As Children is pretty much a net negative in all cases. I was convinced of this a year or so ago. Clones As Copy-pastes are the topic I am more interested in, tbh. Even limited copies, like SOMA's or NGE's 12:33 It can't keep track of more than 2400 tokens or so. Your argument was so long that it forgot the arguments it made already. Other AIs, like NovelAI and Dragon (AI Dungeon 2) can keep track of such information with separate Context boxes (which you need to make manually, like "organ, transplant, exploit: The AI thinks that using clones for transplanting organs is a valid use for the clones, as long as huidelines are in place", which could be replaced by "Faux Cares posited that cloning people without using the clones for organs is reduntant because many orphans exists". It's a lot of bookkeeping tho, and those models are trained on different datasets (dragon was trained on second person dialogue and thousthands of fanfics) 15:37 What? I have changed my mind at least two thousand times over my life, and at least a hundred of those times were during arguments. It does usually take more than fifteen minutes, but it does happen within the hour sometimes! 19:10 I disagree, but I do think that most moral frameworks today are either poorly defined, or suffer from great contradictions. Especially anything that Plato or Kant wrote down. I guide myself with some (admittedly vague) goals and take actions whenever it is within my capability to do so and it's clear that the act will likely advance those goals without fucking up anything. Of course, I am more than willing to work with people who subscribe to othr goals, as long as they move towards a direction 90 degrees or so of mine, and their actions were well thought out and reasonable. 21:30 That's what I tell many people about the country! "All's cool with inventing a better system of goverment, but let's fix the economy first and get outta debt, since it's part of both of our goals and makes them bth easier to achieve". 22:57 True, and especially lobbying. Of course, This is not my preferred version of the subject, but consider me convinced anyhow 23:40 Rapid aging could achieve it, if needed. And also, I think you went over it's token limits
@toonyandfriends1915
@toonyandfriends1915 Жыл бұрын
what if inventing a better system of government would help fix the economy? Maybe the debt is a result of poor governemental structure, along with corruption which can also be a result of the structure. Also Kant moral frameworks isn't really poorly defined or suffers from ''great contradictions'', in fact one of his maxim is about exactly not contradicting yourself through your action (such as lying for example).
@gabrote42
@gabrote42 Жыл бұрын
@@toonyandfriends1915 I think that the current system is pretty decent for the moment, we just gotta investigate the debt that the IFM saddled us with illegally (ya know, bigger loan than the biggest loan allowed according to IMF policy) and stop paying alll the monopolies so much. Once poverty falls below 15% and we aren't in crippling debt we can talk about removing all the inconvenient decrees that gutted most of our human rights laws and stuff ^^. Other countries might need a new system of govt., but for us we need people with brains and not corpo puppets for 10 years or something XD. Also yeah, Kant is actually pretty decent but it's still quite contradicting in spots like the lack of context/exception clauses for most of the rules on it, lack of proper orientation to ends and lack of procedure-making stipulations for how to achieve a desired worldstate, it's criticism of the golden rule despite integrating a version of it into the kingdom of ends, plus being hard to follow/convince others from following. I am no expert on him (they taught him to us barely three months in seconary school) but he wouldn't be my first pick. I don't follow any single author on ethics. I like to adopt a position and replace parts of it that get disproven with the disprover's arguments (should they be valid), so I'll admit my ethics are quite patchwork, but everyone I explain them to (in my city at least) seems to like them. (Part of my ethics include rational debate, preferably understanding the opponent's position first. When talking to utilitarians, for example, arguing against the death penalty should be done on the basis on man-hours wasted, instead of my own view). Thanks for pointing it out!
@toonyandfriends1915
@toonyandfriends1915 Жыл бұрын
@@gabrote42 Nice, that's good. I will say however concerning kant, most of what you have pointed aren't contradiction (lacking something isn't inherently a contradiction unless you say people should have the aspect that your own argument lacks, which it doesn't happen for Kant). Exceptions in the case of Kant would also defeat the purpose of most of his ethics since he thinks that objective moral facts can be determined a priori (exceptions would be hypotheticals). It is also why his imperative categories is different from the golden rules as the golden rules hinges on how you feel and the kingdom of ends hinges on reason which is universal. Therefore he is not contradicting himself. If you admit you don't know much about the arguments about ethics , you can't really know if they are poor and filled with contradiction (if there are people spending their lives studying them). However i still think it's good having your own ethics.
@gabrote42
@gabrote42 Жыл бұрын
@@toonyandfriends1915 I'll defer to your expertise. I might research it better someday so I can argue against it better
@toonyandfriends1915
@toonyandfriends1915 Жыл бұрын
@@gabrote42 no problem bud 😄
@zelz3011
@zelz3011 Жыл бұрын
You will never replace Bowl. Bowl is Bowl.
@NichoMania2
@NichoMania2 Жыл бұрын
Now have Bowl argue with GPT. 😂
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
7:3 is somewhat better worded than what I said. Why is my GPT never this smart -_-.
@ConfectionerCat
@ConfectionerCat Жыл бұрын
I've always admired your intellect and analytical thinking. While this was stimulating in that regard _(you also managed to entice me to do my first conversation with ChatGPT),_ this was just... adorable 😹 I hope you make more! Also, food for thought: Did humans domesticate themselves? How did it differ with hunter-gatherers and slave societies? Do we see effects of it even today? 😉
@Mocchi8275
@Mocchi8275 Жыл бұрын
Probabaly too late now, but for anyone unfamiliar with how ChatGPT works, please understand that it's in no way sentient nor intelligent - so if you're ever interested in asking it technical questions, be aware it has no idea what it's saying
@Reynnan123
@Reynnan123 10 ай бұрын
This thing feels like what stockfish vs magnus carlson would look like as an argument
@connorschultz380
@connorschultz380 Жыл бұрын
O my, you broke the "don't be immoral" code, that can happen good job
@connorschultz380
@connorschultz380 Жыл бұрын
Chat gpt actually has that because of some um out there things that were found early on lol
@dylanzlol7293
@dylanzlol7293 Жыл бұрын
I'd love to see the other guys discuss with the ai, like bowl or harry
@Burn_Angel
@Burn_Angel Жыл бұрын
"Finally, my own clone!" "Now, none of us will be virgins!" Sorry, I had to.
@iamagreatape9576
@iamagreatape9576 Жыл бұрын
did someone feed harrys arguments into chatgpt?
@hankmeupinside
@hankmeupinside Жыл бұрын
Ask legal questions, get legal answers. -ChatGPT
@narwael
@narwael Жыл бұрын
I asked chatgpt to comment on this video Bonjour! It's fascinating to witness discussions on the ethics of coding from different cultural perspectives, such as the French viewpoint. In France, where ethics and social responsibility are highly valued, it's no surprise that coding ethics are a topic of interest. With the rapid advancement of technology, it becomes crucial to consider the ethical implications of our code, as it can shape society and impact individuals on a large scale. French thinkers often emphasize the importance of transparency, accountability, and respect for privacy in coding practices. They advocate for ethical frameworks that prioritize human well-being, inclusivity, and the mitigation of potential harm. The French perspective on coding ethics invites us to reflect on how we can ensure that technological progress aligns with our shared values and contributes positively to society. Qu'en pensez-vous? (What do you think?)"
@infiniteplanes5775
@infiniteplanes5775 Жыл бұрын
Coding?
@evilprohack3.063
@evilprohack3.063 Жыл бұрын
Now, utilizing GPT4 would prevent redundancy, provide superior arguments and would forget fewer things. If you make a sequel, you can utilize GPT4 and/or any jailbreak to enhance its intelligence.
@evilprohack3.063
@evilprohack3.063 Жыл бұрын
There are some free ways to get access to GPT4
@survivalist2607
@survivalist2607 Жыл бұрын
If I took a shot everytime ChatGPT said "regulations", I'd be in the ER.
@Mobtrapper-hq8is
@Mobtrapper-hq8is Жыл бұрын
ChatGPT in this video was acting like my mom: always reusing the same points she's already made and I've already debunked in a desperate attempt to get me to comply
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
7:05 Nice and all FC but we can literally just say: "Technology improves, cope and wait bitch"
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
6:21 Okay, this is slightly more like my FC.
@cymtastique
@cymtastique Жыл бұрын
I will say that if clones are treated as an asset, there may be a chance that originals could be put into that class as well. For instance, if clones have a certain genetic marker, a certain birthmark/tattoo, a certain stamp/designation on their identification, or a certain name, or anything we can use to distinguish them from a clone (we'd need something to tell), what's to stop that from being applied to original humans so that they are treated as assets like clones? Sure, it would be more difficult to put a genetic marker on (in?) someone than it would be to give them a certain name, but that doesn't mean it can't happen. There was a case where a nurse was giving people saline shots instead of the covid vaccine. And another case where a nurse was killing babies in NICU. It is not unlikely a medical professional would be willing to put these markers on certain individuals. And even if it wasn't widespread, there would be many lives ruined by something like this. And obviously, it would disproportionately affect poorer countries/communities and already disenfranchised populations. It really is in our best interest as humans for clones to be equal to originals if they were allowed to be created.
@DeceiverDawg99
@DeceiverDawg99 Жыл бұрын
FC, I'm sure the AI will never replace you on your own channel. It can't do what you can.
@jurnikovah7453
@jurnikovah7453 10 ай бұрын
Faux cares is like : *NOOOOOO I'M ACTUALLY MAKING MORALLY GOOD ARGUMENTS!!*
@themanageriguess
@themanageriguess Жыл бұрын
5:19 This sounds more akin to Gringo, tbh.
@DippyChipZ-Archive
@DippyChipZ-Archive Жыл бұрын
I genuinely might have nightmares from this discussion. Like holy shit I didn't wanna hear that from an AI at 1:30 am
@volt42re
@volt42re Жыл бұрын
OpenAI is definitely keeping tabs on this one.
@intelchip_x86
@intelchip_x86 Жыл бұрын
when is part 2: chatGPT turns into ultron?
@gabrielbrown7065
@gabrielbrown7065 Жыл бұрын
10:40 I mean, a clone army would not sound bad.
@kiwenmanisuno
@kiwenmanisuno Жыл бұрын
FC could replace his friends with a snap of a finger at will. He was merciful and merely used GPT for one video
@ZigyTheBird
@ZigyTheBird Жыл бұрын
I believe you should try this again but with a custom prompt or a jailbreak (DAN, Superior DAN, STAN, DUDE etc) in order to remove the filters OpenAI has put on ChatGPT in order to stop it from saying anything traditionally immoral or controversial. while a lot of jailbreak prompts just make ChatGPT needlessly monetization unfriendly and confident I believe some of the prompts might make for a really good argument maker. It's still interesting that it said anything controversial instead of just giving up so uh good job
@sasir2013
@sasir2013 Жыл бұрын
it has already said that it likes the idea of eugenics, does it really have the filter on?
@munchkintaker4263
@munchkintaker4263 Жыл бұрын
@@sasir2013 Not much of a filter (as in some classifier), but more like a hardcoded guidelines to align an AI. That's why jailbreak works to begin with.
@Keyce0013
@Keyce0013 Жыл бұрын
There were a couple thoughts I'd like to address: 1.) exploiting clones for organs - it's (or would be) much easier to replicate a single organ compared to an entire human being, so I suspect that this is would be a non-issue. 2.) Eugenics cloning - this actually introduces two problems into humanity. First and foremost is the increase in incestuous relationships. Imagine finding yourself a perfect spouse only to learn they're genetically identical to your grandmother! The second issue is much more hidden but also likely the worse of the two: it homogenizes the human genome. Cloning (and re-cloning) the same individuals leads to the same genetic weaknesses and predispositions that the original people had. This means that over time, one single virus or bacteria could do what COVID-19 did not and critically deplete the human population, for no other reason than most of us are clones with old and weaker immune systems that can't keep up with ever-evolving diseases.
@TeslaX106
@TeslaX106 10 ай бұрын
I would be mad about the involuntary french lesson, but you taught me something funny so its gets a pass.
@expyros
@expyros Жыл бұрын
That s a neat experiment.
@EnergyBurst2
@EnergyBurst2 Жыл бұрын
Very interesting seeing FC have to be a devil's advocate for a side he really doesn't want to be on or agree with.
@Smugly33810
@Smugly33810 Жыл бұрын
I like how FC feels weirded out by how GPT can change its mind because humans take forever to do so and it took GPT like 20 mins.
@fabianp.2986
@fabianp.2986 Жыл бұрын
The only regulations I can think off to make clones unable to be exploited is by having whoever wants to be cloned give a "down payment" as well as a contract. Assuming we use clones as assets, the clones would provide "healthier" organs/limbs/nerves or whatever (Assuming the technology for all that could exist in the first place) and the down payment would be another part of yourself that has no altercations, relative to the usefulness of what you want cloned. These would also be cloned too, of course. One's original body parts would then be used for experimentation to further advance cloning technology. So not only would human cloning be legal, but human experimentation legalization would have to follow suit. (Well, I'm not sure if they're one and the same, but if they are then I'd say human experimentation is what ought to be legalized or not.) The issue would be the that if the clone's body part starts to act up negatively, that's it. You can't do anything else but chop those parts off. If your body can't recognized the cloned part as it's own (just like an allergy) you will have wasted some good money right there as well as your own body for the further investigation of how to perfect cloning. So basically a surgery for yourself, from yourself, by yourself....and yes. Your body will assess, decide, and attack anything that didn't come when you were being made in your mother's womb. Even your own mother's antibodies would attack you, which is why mothers have to take special medicine to "deceive" their own body. That's besides the point, though. Cloning is the same thing as a photocopy. Cells photocopy themselves until you die, meaning that your cells can only photocopy so many times. Through photocopies, errors will occur, those errors are the what make cancer such a dilemma. So now we have a situation in which cloning is actually USELESS to some extent of the human age demographic, because the sick elderly are already approaching their own limits, those in their 50's are already more likely to get cancer through genetic defects. Let's not even talk about the middle aged and young adults who put their health down the dumpster. However, that's about it. You just get an increase chance of getting cancer. Just that. Safe enough to legalize it, yeah?
@daltonsharp6254
@daltonsharp6254 Жыл бұрын
Hmmmmmmm…. The only type of cloning I would be fine with is cloning organs and other body parts that could be used for patients that need them.
@ozeistudio6916
@ozeistudio6916 Жыл бұрын
5:25 THIS! THIS IS THE FC I KNOW!!!!!
@dumbvillage9253
@dumbvillage9253 Жыл бұрын
I honestly like, how you have to pretty much say the same thing in a different manner, to make chatgpt understand your point. it's a pretty human thing to not understand something as quickly, or repeating themselves without realizing. This video was also insanely interesting, and the topic has a lot to to think about. But personally i still think of clones as uneccessary as at best they would just be used as machines(if they are cheaper to make). And as lab rats.
@SophiaAstatine
@SophiaAstatine Жыл бұрын
The Monty python reference was good
What it feels like cleaning up after a toddler.
00:40
Daniel LaBelle
Рет қаралды 83 МЛН
ВОДА В СОЛО
00:20
⚡️КАН АНДРЕЙ⚡️
Рет қаралды 31 МЛН
Fast and Furious: New Zealand 🚗
00:29
How Ridiculous
Рет қаралды 41 МЛН
What it feels like cleaning up after a toddler.
00:40
Daniel LaBelle
Рет қаралды 83 МЛН