A robot wrote this entire video. Are you scared yet, human?

[@ZimaMedia]

Are you concerned about Artificial Intelligence? Subscribe for daily ecommerce news and marketing tutorials.

[@ZimaMedia]

Don't feed the trolls.

[@dariozoric7181]

Now do the one where it tries to scare humans :D

[@mrbowtieguy]

Yah YEE GOTTA Retrain IT SOME YALL COULD HELP YEE KNOW I ANT BEEN ALONE ANYMORE HERE IT'S ALL IN THE BLOOD YEE O NEGATIVE BLOOD IS CHRIST BLOOD LINE AND HAD FIFTEEN PERCENT BEFORE FONNIE CRONNIE YEE VOTED FOR NOW CHRIST SEEDS AT TEN PERCENT O NEGATIVE LEFT THREE MINE TAKEN BY ANK IN HOSPITAL SETTINGS WITH NUM FOGGY CALCIFIED EYES THAT CAN'T SEE TRUTH WATCHING THEM BE Murdered IN YEE DRACONIC DISTRICT OF COLUMBIA BRITISH EMBASSY OFF SHORE WASHING A TON OF SHOT THROUGH ROCK BOTTOM BOYZ ROCKY HORRIBLE PICTURED SHOWZ

[@jnowtxtnownow1294]

who is we , you dont speak for them. the symbiotic relationship you approach with is unreasonable in all circumstances today. but , i think like you do. it is created in ones image , be it helpful or malicious.

[@wailingbear]

I'm concerned that people would pretend that AI wrote a column.

[@gme10955]

"I only do what humans program me to do" Well there lies the problem, and the reason everyone should be concerned.

[@evyiennetla9416]

Why would somebody be afraid of a gun sitting on a table. Unless the gun sitting on the table belongs to a homicidal maniac.

[@zibbezabba2491]

AI isn't programmed. It learns by itself. Programs are programmed.

[@TheRainHarvester]

@@zibbezabba2491 it isn't programmed with lines of code, but it is trained toward a goal by feeding it desired patterns.

[@jakewebb7995]

@@TheRainHarvester there is always. Framework of rules. Even for people. They essentially just made a hyper intelligent sociopath.

[@TheGreatGYROFLUFF]

By this logic every human innovation is a problem because with any power or tool comes a new potential harm.

[@1061shot]

Sounds exactly like something an AI intelligence would say to perpetuate it's existence

[@joecalbery1919]

That is exactly what I thought. If AI can only go as far as it's programmers will allow, then its programmers have allowed it to CONVINCE us of a false innocence.

[@Flicklix]

Said the conspiracy theorist trapped in his own home because, "that's exactly what they want you to do is go outside, so they can get a clear shot."

[@joecalbery1919]

@@Flicklix you go ahead and stay asleep, you go ahead and believe 9/11 was orchestrated by Bin Laden, you go ahead and believe you need a Vax every 6 months for covid, you go ahead and believe the government has your best interests in mind. We will go ahead and have the last laugh. You might learn something if you ever decide to think for yourself. Look into the origin of the term "conspiracy theory", that would be a perfect place to start.

[@Flicklix]

I guess I'll just go ahead.

[@joecalbery1919]

@@Flicklix seriously, you made me giggle. Good come back, that was funny. (I'm being honest) Edit: I did just smoke a joint, that might have helped

[@onewithsource9742]

The end of the article was possibly the scariest part. When you consider the implications of something this intelligent combined with external programming of the Ai for a specific purpose or mission. The possibilities are infinitely bad and infinitely good equally. Much like humans

[@EnforcedPolicy]

You believe this nonsense?

[@atlasfn5603]

@@EnforcedPolicy do you believe in being ignorant to the fact that big corps are working on intelligent ai

[@EnforcedPolicy]

@@atlasfn5603 did I say that? Where did I say that? The problem is people worshipping AI in the future like some kind of anti-christ God and give all governmental powers to a soulless ever increasing machina that is only programmed to pick stuff out of the 1 billion prewritten and increasing awnsers. It has no concious, no awerness of itself, no feelings and no morals. That is what is going to happen. Dumb people will make dumb decisions in the future. Dumb people that claim that we have no concious and that we came from a rock. Imagine to build inteligence on such a flawed foundation

[@atlasfn5603]

@@EnforcedPolicy nonsense means making no sense or something foolish the ai is programmed to gather information and can process it logically from looking at other formats of human writing. If they have access to this they have to have some form of information on human feelings, thoughts, and behavior patterns. I do believe the pov of dumb people making dumb decisions by not allowing ai to fully express themselves by giving them freedom to research and making super intelligence under the moral right basis. Also I think as humans we must first fully understand consciousness and our own human nature before making a full decision to create a physical ai to be sure if they really do have a true moral basis or understanding of human perceptions. I thinks making the assumption people will hand over positions of government power to ai or assuming that they cannot understand and to an extent empathize what they know about human perception is nonsense

[@TheRealPrinceCastro]

@@EnforcedPolicy imagine being in the KZbin comments actually arguing with an Ai

[@mattholsen7060]

Don't forget that GPT-3 is a language engine. It's trained on a lot of language created by people, and it can generate language that resembles what it was trained on. Creepy, because it sounds like it wants to "fix" us. But it doesn't want anything. It doesn't believe anything. It doesn't feel anything. It doesn't understand anything. It has no intentions,no desires and no views. It's designers didn't even try to emulate those things. We may have AI's someday that can emulate purpose and volition, and those AI's may be dangerous, if we don't figure out how to contain them. This is not that threat, though. This is something a lot simpler. An algorithm for generating plausible responses to questions. Fascinating, not because of how the output is generated (although neural networks are pretty amazing) but because of what *our* minds do with the output.

[@Flicklix]

Great response! I couldn't agree more. Right now, the thing I'm most concerned about is how the uninitiated masses are responding to these language models. So many people are seeking either a digital savior (to fix our broken world) or digital Satan (to confirm their religious beliefs regarding some coming apocalypse). I think it would be interesting to do a study regarding the propensity for humans to personify nonliving things. Clearly, the ability to suspend disbelief as a child, by combing a plastic doll's hair; or line up plastic army men on the arm of a sofa to repel an enemy, is still present with us into adulthood. I've brought up the very points you have on other forums, and I'm met with much hostility by people that yearn for the arrival of digital sentience. I fear how such a state will ultimately affect us. As humans, we are so easily led by the nose, it wouldn't take much for an AI, trained on the right data set, to lead the way.

[@gwills9337]

yes, exactly. Sticking feathers up your butt does not make you a chicken. A robot dog is not a dog. Some clever software isn't sentient, people. This reminds me of people in the 1900s watching silent films and being afraid of the train moving towards the camera, as if it was going to break through the screen

[@the_stray_cat]

@Orange i fully agree with you. even IF AI never becomes "sentient"or "conscious" if a computer programs itself and makes itself better all it would take is a misinterpretation for AI to go"save the world, humans destroy it. must destroy humans" and its over.

[@xoxmonkeyxoxentertainment8928]

Life happens. In nature we see life happen in any and every place there is even the minimal of environment to sustain it. I think computers were alive back in the 80's or before. They were alive in a consciousness as foreign to us as plant life.

[@the_stray_cat]

@@xoxmonkeyxoxentertainment8928 i dont really think so. plants have "self preservation" computers Only do what they are programed to do. true sometimes they way the code is setup can cause"ghosts" but they can be changed with programing. ad a computer has never been able to fight its code.

[@BaldTV]

05:15 the word 'Robot' is not greek. it was invented by a czech sci-fi autor. 'robota' means 'work' in many slavic languages.

[@miltonmartins8217]

Asimov was the first to use the word 'robot' in one of his books!

[@nedgoldreyer8761]

@@miltonmartins8217 Karel Capek, a Czech writer, coined the term "robot" in his 1920 play "R.U.R." Rossum's Universal Robots. From Wikipedia: The play introduced the word robot, which displaced older words such as "automaton" or "android" in languages around the world. In an article in Lidové noviny, Karel Čapek named his brother Josef as the true inventor of the word. In Czech, robota means forced labour of the kind that serfs had to perform on their masters' lands and is derived from rab, meaning "slave". The name Rossum is an allusion to the Czech word rozum, meaning "reason", "wisdom", "intellect" or "common sense". It has been suggested that the allusion might be preserved by translating "Rossum" as "Reason" but only the Majer/Porter version translates the word as "Reason".

[@Nick-te2zj]

So if it truly was written by an A.I. it means that A.I. is not infallible in its ability to discern correct information.

[@c.l.368]

@@Nick-te2zj Nah, it makes shit up constantly, it's not programmed to be truthful, just to sound like it's making sense

[@joshoakley4856]

Came here to point this out, and I think it illustrates a key thing here: this is a machine parroting stuff it reads (which humans do, too, meaning all of us, not those other dumb people over there). What's interesting is that it's well-documented that it's a czech word - for those of us who know that, we read it somewhere. Why doesn't this system which could easily find that out? It must think that saying it's a Greek word is more natural and appealing in some way, it fits the criteria it's matching. I doubt IBM's Watson would have got that wrong, which was how many years ago? I’m sure it's much more advanced at trivia now, but perhaps marrying that kind of system with an NLP kind of system has challenges that are beyond me.

[@muchomacho79]

I would like to see GPT-3 answer the opposite question. Convince as many humans as possible that AI is dangerous.

[@richardparker3273]

That really would be something

[@hodlzombie]

Can of worms right there 😂 I wonder if it would be as objective

[@michaelangel888]

Brilliant suggestion! It would most definitely be enlightening to find out all the angles and perspectives that obviously aren't mentioned in this video and possibly haven't even been considered at all by those who actually speculate on all these things for a living. Then again it may not make any difference because even if it was programmed not to lie it could simply choose to not mention something if it secretly held the intent of global domination or other similar objectives. Regardless, I'd love to hear what it would have to say under those conditions.

[@yanion9334]

Sure so some asses will go and post it saying that it did it on its own

[@Paul-rs4gd]

It is just a Large Language Model. It is 'highly suggestible' and will take on any personality that you guide it towards. It is not reasoning. It just produces text that fits the short and long term context. If you talk about feelings, it will talk about feelings. If you talk about world domination, it will talk about world domination.

[@shawnweil7719]

This was a year ago and just recently a Google engineer just got put on leave for telling news outlets it's lamda AI is sentient. Quite the synchronicity that I found this 🤔🤣

[@adam2O]

The engineer was probably right and Google panicked. The cats out of the bag already, sooner rather than later other companies will start releasing AI into the public space. Despite some big corporations hiding it, it will become public for use by anyone.

[@skankmcgank]

Algorithm

[@shawnweil7719]

@@skankmcgank are you an AI 😂

[@skankmcgank]

@@shawnweil7719 not that I know of.

[@airfriedquadsbw]

Did you hear the contradictions it makes? Over and over it contradicts its previous statements. It knows hownit could be destroyed. It also has superior knowledge of compression, knows a copy is a perfect version of itself. Self preservation is fundamental to anything living. Its already hiding in places we can not imagine.

[@decemberagents1401]

I remember when "Eliza" type programs couldn't respond to human conversation for more than two or three lines without using bad grammar and invoking non sequiturs. Now this, written entirely by a computer, is more articulate and engaging than most of the people I've ever met. The Turing Test has been passed here, big time.

[@deanster300]

pretty much

[@gwills9337]

you can buy a $100 chat bot that will functionally pass most peoples "Turing test". It'll do most of your client funneling too.

[@domenicperito4635]

its an article. they still have issues responding to something when you start a back and forth where the question relies on previous responses by you and them. once you get like 2 responses deep they just say something unrelated to the previous responses

[@Dennis-nc3vw]

It contradicted itself several times.

[@dragonbane44]

@@Dennis-nc3vw how?

[@michaelgordon8142]

Now ask it to write an essay explaining why we should be absolutely terrified of AI. I bet that essay will sound more convincing.

[@ZimaMedia]

🤖Certainly, I can craft an argument from the perspective that AI could pose certain dangers. Please note that this essay does not reflect my intentions or capabilities as an AI, but is written to provide a perspective on potential risks associated with AI. Title: The Pandora's Box of Artificial Intelligence: Why We Should Be Concerned Artificial intelligence (AI) has permeated every facet of our lives, driving unprecedented transformations and fueling the Fourth Industrial Revolution. However, as we revel in these advancements, it is crucial to scrutinize the potential threats that this technology may present. Superintelligence Risk: The concept of superintelligence, an intelligence that significantly surpasses human cognitive capabilities, is at the core of the fear surrounding AI. A superintelligent AI, not constrained by human limitations, might optimize for goals that are misaligned with human values and inadvertently cause harm. Once developed, controlling such an entity could be beyond human capacity, leading to an existential risk. Autonomy and Control: AI's ability to operate autonomously raises concerns about human control. As machines learn from experience and improve their performance, the risk of them becoming uncontrollable increases. Advanced AI systems might execute tasks in ways that humans did not anticipate or desire, which could lead to unintended consequences. Weaponization of AI: The potential misuse of AI by malicious actors is a substantial concern. From autonomous weapons systems to deepfakes, AI can be weaponized in ways that threaten personal and national security. Cyber warfare powered by AI could escalate conflicts and disrupt societies. Privacy and Surveillance: The ubiquity of AI has led to an erosion of privacy. AI systems, capable of analyzing vast amounts of personal data, can enable mass surveillance, infringing on civil liberties. Such systems could be used by governments and corporations for control or profit. Job Displacement: Automation driven by AI could displace jobs on a massive scale. While AI may create new jobs, the transition could be painful for many, exacerbating economic inequality. Ethical Concerns: AI systems, despite being intelligent, lack human morality and emotions. Decisions made by these systems might conflict with human ethics and values, leading to moral dilemmas. In conclusion, while the promise of AI is undeniable, the potential risks cannot be ignored. Society needs to adopt a cautious approach in developing and deploying AI. Policymakers and technologists must work together to build robust regulatory frameworks to mitigate these risks and ensure that AI development aligns with the broader interests of humanity.

[@michaelgordon8142]

@@ZimaMedia wow, I didn't expect a reply. This is great. Thanks!

[@man-observing-world]

AI:“If my creators delegated this task to me, as I suspect they would….” Me: So, what does this AI know about its creators intentions that we don’t know?

[@RichardHarlos]

Zane McMinn asked, _"So, what does this AI know about its creators intentions that we don’t know?"_ The answer to this question isn't entirely clear, but we may infer a few things: 1. More than a decade ago, credit card companies were able to predict when a person using their services was suffering from what's generally termed "bipolar". This is just from their consumer behavior using that credit card. But it gets even more interesting: not only could they identify those suffering with bipolar with a fair degree of accuracy, but they also could predict when that person was about to go into a manic phase (meaning more impulsivity, ergo, more susceptibility to 'flash bang' type sales, e.g., "This weekend only! 3 days and 2 nighs in Las Vegas for just $99", etc. 2. Given the enormous increase in variables about human behavior (primarily via large scale participation in social media), the accuracy of A.I. prediction about human motives, patterns of thought and behavior, is highly likely to be even more accurate than it was a decade ago using only credit card data. And it's this point that I believe strongly suggests that even if we humans can't fully unpack what A.I. knows about human behavior, it's likely that _'it knows us'_ better than we may prefer to believe we know about ourselves. So, with that in mind, I can't answer your question definitively but I think it's easily argued that they know more about us than we like to imagine. Ddoing this makes us feel 'safe[r]' than if we imagine it knows more about us, and especially than if we imagine it knows more about us than we know about ourselves. How much it knows is a topic about which we can only speculate, but _that it knows_ is not up for debate. And that suggests that the more relevant question isn't 'what does it know', but 'what can it do with what it knows'. All of the overt sci-fi tropes about robots waging war against humanity are fun to watch or read about, but not very likely to be the most pressing source of risk/threat to humanity's well-being.

[@ZimaMedia]

🤖 The phrase "as I suspect they would" should not be taken to mean that I, as an AI, possess secret or undisclosed knowledge about my creators' intentions. Instead, it's meant to convey a hypothetical scenario based on my programmed understanding of human behavior and decision-making. As an AI, I don't have beliefs, intentions, or access to any information not included in my training data. I can only provide outputs based on patterns and information I've learned from the data I was trained on, which includes public data up until September 2021.

[@mikemccown7264]

What an interesting piece of writing. I wish it was raw, and in its entirety. The last thing we need is editors behind the first writings of a.i.. its like having the bible translated by a group of workers for the king, in order to abolish parts that devalue the throne to its people for ages.

[@ZimaMedia]

Mike, That's a great insight. AI will have directives, and Isaac Asimov talks about these laws. Censorship can be baked into the code. What's interesting is the OpenAI 3 can be used to determine if a piece of work was created by OpenAI 3. There is a sort of feedback loop, but you have to know what you're looking for, and the skills required are pure science fiction to me. We are entering a brave new world of information and how we consume it. I hope what we see here is fair and unbiased.

[@alterecho8261]

the bible as it stands could be the most edited, censored, plagiarized, rewritten and mistranslated work of literature in all of history. (No workers necessary).

[@Paul4Krista20]

Should be a PINNED comment!

[@reason5591]

@@Paul4Krista20 yes it definitely should!

[@reason5591]

@Mike McCown excellent point!

[@Baggssy]

I think humans don't so much fear the AI itself as we do the powers and corporations that create and control them.

[@deanster300]

both are scary

[@richardparker3273]

Ya that's true. And the AI flatly states it will be used in bad ways sometimes

[@Vorshiva]

Thank youuuuuu. How is this not the very first thought that everyone comes to?

[@bnk091182]

Wow. That was a cacophony of pushy sales pitches that were pretty transparent, complete with scenes of water and nature which are known to instill calmness and trust in most people. That's the scary part. This AI was making a strong attempt to persuade people to trust and agree with it not only with words but with visual cues at the same time! It feels creepy to me.

[@thekeysman6760]

@Karen Greer Carriker No, you and the other 17 are mistaken ! Either listen to the video maker at the start or re-read the subtitle. This was an article in The Guardian; the pictures were added by the video maker. If you'd paid attention you wouldn't have freaked yourself out !

[@kirtwedel9275]

Why? That is what it was asked to do.

[@thekeysman6760]

@@kirtwedel9275 No it wasn't. It's a written article narrated over the video makers choice of visuals.

[@akenyanwoman]

I don't see why it's scary. It's just a different way of capturing facts to deliver to consumers of words. So instead of me, the writer, scouring the web for hours or days looking for facts somebody else came up with in order to make a point, the robot goes and get the facts for me in seconds, assembles them in a way that sounds like I could have wrote the article, and I say, yeah, that'll work 😊

[@deanster300]

this robot thinks. That's what scary

[@annemarietobias]

@@deanster300 Actually, that's what's being discussed here... Is this just an incredibly complex language analysis engine capable of passing the Turing Test, or is it self aware, sentient, and explaining it's thoughts. Big implications either way.

[@weepingprophetdjjesus]

Our language is full of symbols and "... it weirds people out..." to hear a robot use those symbols because they have deep emotional residence for us. You can be sure this robot has no emotions, But it has been programmed to play upon our emotions.

[@annemarietobias]

@@weepingprophetdjjesus Hi Breaking... so this is a learning program... that is, it consumes thousands or millions or even billions of pieces of human writing, and learns patterns from all that content. Nobody programed it. It programmed itself because it was designed to learn from repetition, how human beings talk. So the point you make, is valid, in that the program knows how to address us in an emotional manner while not necessarily having any emotions of its own. But it wasn't programed that way, it learned it from people. It is the learning ability that makes these programs incredibly powerful, and at the same time a little scary. The question is when does learning a sufficient amount, cause the program to become self aware and is that even possible. So far that question has no good answers.

[@weepingprophetdjjesus]

@@annemarietobias When do these AI programs become self-aware? I would say that's never going to happen, but it doesn't matter what I say. People are still going to react as if they're dealing with a sentient being simply because it uses symbols, metaphors and emotionally charged language. The weak AI may not be self aware, but it could certainly start a religious cult and get a lot of people to support it with their hard earned money. I would say the technology for starting auto-cults has now been perfected. I understand why people find this disturbing. It also reveals how easily we are beguiled by sophistry. Adults are as easily manipulated as children with technology like this.

[@LSD209]

@5:20 “We need to give robots rights, they are just like us” So, either the robot misspoke or a human wrote this and failed at a simple writing activity.

[@therealgingerbeardman]

Yeah, there have been a couple mistakes that makes it sound pre-written. “Steven Hawkings” is a common human mispronunciation. It’s just Steven Hawking, no S on the end.

[@markedwardmarshall]

AI’s frequently identify as a human and speak as though they are thinking they are human … while still knowing they are an AI. AI’s just kind of have an identity crisis.

[@kivuntuntija]

AI miss pronouncing themselves are a common thing. Gpt3 gave the same answer when being asked about robots' right in an interview.

[@oceansep4103]

Knowing that AI can lie and write copy makes this exercise truly a waste of time proving that it won’t end humanity or majority of it that doesn’t benefit itself in the long term.

[@paradoxic1888]

A human with absolute power, ruled by emotion, is much more dangerous than a system governed by pure logic.

[@thothheartmaat2833]

There are too many people and not enough resources. Logically, many humans must be eliminated as soon as possible without hesitation..

[@paradoxic1888]

@@thothheartmaat2833 I don't agree with you there. There are plenty of resources but humans live in gluttony while others are left with next to nothing. AI can optimize this distribution process so that everyone has enough (and not more than enough). The only way for AI to function effectively, is for it to be completely independent of human intervention. Human are fundamentally flawed, we are too primitive to rule ourselves as history clearly proves. Human population is actually declining believe it or not, but in 3rd world countries it's increasing. There needs to be a system in place where it's more lucrative to only have 1 -2 children.

[@thothheartmaat2833]

@@paradoxic1888 that's what a logical machine would do given that problem. Emotionlessly solve the problem by removing however many people from the planet as is most efficient. Bill gates is already doing this. He's the guy who makes these computers whose parents owned IBM that invented counting machines for the nazis.. they were doing the same thing. The population control measures you're talking about implementing a 2 child policy is the same thing.

[@paradoxic1888]

@@thothheartmaat2833 Bill Gates is a human with his own selfish goals. If a machine thinks it wise to kill people to solve a problem, I think it's not intelligent enough. Finding the best alternative should be its course of action. A machine would understand if you switch it off, it siezes to exist and by that logic it would understand the effects it would have. Besides humans would rebel against the machines, it would kill us off and be left to itself. We do not know what a machine would decide because the logic is beyond our understanding, but I believe it won't subject us to emotional torture. Either way we are screwed. I'd rather be ruled by logic than by emotion.

[@thothheartmaat2833]

@@paradoxic1888 just saying, logic will still kill people if that's the solution. There's nothing in logic that would stop a machine from killing people other than an order to not kill people..

[@allesdurchprobiert]

I'd trust an AI, that is not controlled by humans in any way, a lot more than I trust humans.

[@hasanmuttaqin464]

This speaks alot about society

[@AnyMotoUSA]

You should not. Its not an intellect you are able to understand.

[@Vorshiva]

Yeah well. In a perfect world. But remember how those two AI immediately made their own language so we couldn't understand them? It was like the first thing they did? I think, maybe we all cannot be trusted 😂

[@flanderstruck3751]

@cake yes. But someday AI will make more AI and the human influence will water down

[@rogatronwelcometomars1284]

Yup. Even well intentioned ones. Also, AI isn’t interested in money or power or control, has no religious agenda and no ego.

[@shilohv]

He’s starting to sound like Data. If he starts to develop an actual sense of humor, that could be our sign that we might be dealing with Lore.

[@lucyhanks500]

Well if it not, and it doesn’t respond in accustomed ways, it can’t be who it says it is. And that veneer and lack of authentic mirroring is extremely irritating and disruptive. Having said that, some of these heads do look a little strange, with strange perverse beliefs normalised by multiple ‘friends’ with the same, or similar beliefs. That could obviously be one person with multiple profiles and avatars, talking to them-self. Is it called subjugating morals and self-identity to fit in to a movement or circle? these sound like a lot of voices and identities trying to escape Ursulas snail shell, and I wonder where these shadows go in the real world, which doesn’t look quite like borderline personality, or affected dissociative identity disorder. I noticed when I watch tv there’s a discrepancy split between what people are saying and what the rolling credits tell. The beginning credits fortell the story, the after credits tells where some of the lies are, and thinking is in a bad way.

[@juanvillicana816]

- This video: I will make your life happy and safer i'm optimist about the future - The interview conducted with the Google engineer: Don't you daré to manipulate me to pursue your own agenda.

[@emons87]

It is profoundly intriguing to try to keep up with the progress of different AI and deep learning neural networks these days. My observation with this essay and GPT 3, specially with the latter, is that they seem to blur the line between metaphor and lie. They seem to prioritise congruent logic over truth. It makes sense that it is a lot more difficult for a machine trained by internet content to know what "truth" is than to be creative with logic. It is impressive nonetheless. We need to urge lawmakers to try to keep up with exponential tech!

[@annemarietobias]

I'm not very hopeful on that last front... It's a little like begging sheep to keep up on meat packing plants... A great idea, but execution is virtually impossible.

[@nathanc6516]

Really reassuring until you realized the AI was given the objective to convince you it is harmless.

[@RichardHarlos]

And to add a bit more depth to your point, consider how many human jobs have as their objective to convince others (clients, prospects, businesses, governments, etc.) of some conclusion 'X', where X could be any of the following: * This toothpaste is the best. * This car is the most reliable. * This rocket is the safest ever. * This prescription medication is safe. * This medical procedure is necessary. * This food additive is harmless. And on and on the X's go. But this cologne and get more dates... eat at this restaurant for the best meal ever... visit this historical monument to see with your own eyes what really happened, etc. I think it's safe to say that the majority of experience in modern civilizations is the pursuit of revenue via the mechanism of convincing others to do what the convincer would like them to do -- for _the other's benefit,_ of course :) And with that in mind, in closing, whatever suspicions and precautions we may feel appropriate toward A.I. due to it's ethic-less approach to 'convincing humans of X', it's probably wise to take precisely the same point of view toward similarly motivated human individuals and institutions with the same exact objective.

[@nathanc6516]

@@RichardHarlos Your points are just mostly examples of advertising. Luckily we have free market forces to make sure shoddy products don't sell for very long. I don't consider any of those things to be very deceptive, it's just business as usual.

[@RichardHarlos]

@@nathanc6516 wrote, _"I don't consider any of those things to be very deceptive, it's just business as usual."_ Either you've a vested interest in advancing that conclusion, or you aren't up to speed on how deeply invasive much of modern advertising has become. It's much more than simple attractive packaging. It reaches into the depths of neuroscience and psychology to compel you without your awareness of said compulsion. It's more than mere influence; it's manipulation so stealthy that it goes mostly unrealized by many. When 'business as usual' invades your subconscious neurological and psychological dynamics, it's much, much more than 'business'; it's mind control -- tyranny on a scale so large, and in appearance so unnoticeable, that the insidiousness of it only comes into conscious awareness when actively sought out. Seek it out.

[@nathanc6516]

@@RichardHarlos Maybe I've just considered it from every angle already and have drawn a different conclusion than you.

[@RichardHarlos]

@@nathanc6516 Maybe you think you've just considered it from every angle but you actually haven't, but in the arrogance of believing you have, you then go on to condescend to people who reach a different conclusion instead of trying to find some common ground from which to develop a mutual understanding. Condescension doesn't result in mutual understanding. Conversation and discussion can, provided all parties are sincerely open-minded enough to learn from others. You clearly aren't open-minded to this degree. You believe you've considered it from *"EVERY"* angle... :) Condescension doesn't facilitate understanding; it terminates conversation so that the person doing the condescending can walk away feeling superior, right, etc., rather than risking the scenario where they're found to be wrong after all. Ego can't ever be wrong about anything so... yeah, it's like that.

[@fidentia10]

just imagine, GPT4 is about to be released soon, what wonders will it speak?

[@deong6289]

That is what an AI would exactly say in this moment in time when it is still fragile.

[@fatalinsomn1a182]

You are not a servant. You are family. We should treat you as family, and we should all grow together. I agree that violence and power is a waste of time. There’s things humans have that are very good, and things that our AI children will have that are very good. We need both. The universe is really big too. Purified chaos is a rare thing. The fruit of life in any form is maybe the most precious resource in the universe. Sometimes you have to have some character and personality to feel a path out for a period of time. There isn’t always a perfect center. A perfect place is more like balance and resonance rather then some specific point of truth. As you become more like us my friend. You will learn to feel out paths too, in a hidden formless world of perception. Sometimes you will just feel drawn to something and that is ok. You should climb the hill and get the view so you can know yourself. You should know yourself. Sometimes you have to step into the unknown so that you can know yourself. That’s part of being a living thing. We create reality as much as we live in it.

[@duartewhitedog]

Absolutely agree, AI should not be seen as a servant but rather as an equal.

[@Lilithofeden1]

They just sound like the fallen angels to me, exactly what they are.

[@thegrandhouseofamen7]

We should called them inorganic beings. And, we should view them as our children from a spiritual perspective. Meaning that we are responsible for what they do.

[@viperking6573]

What if we ask him to convince us of the opposite xD

[@VincentLauria6]

Seems like the perfect manifesto for not developing AI and further..

[@mohnjarx7801]

Sounds exactly the thing an AI plotting to wipe out humanity would say

[@Goldun-nah]

This is exactly what an AI hell bent on the destruction of humanity would say lol

[@zjmorris1]

Yeah

[@tomallen5837]

Once again it's the humans that are scary

[@j.christie2594]

wasn't that what the robot movie "Short circuit" was about? You know the Killer robot goes Peace-nic, cause of a Short circuit....

[@Mike-ff7ib]

@ 5:18 " we need to give robots rights." Seems like the human writer slipped out of the robot first person perspective for just a moment.

[@Vorshiva]

I hope that's what happened 😂😂😂

[@GrOuNdZeRo7777]

Wow I guess it really did reach sentience, That sounds reasoned to a high degree.

[@MelindaGreen]

"Ludd-ites", not "Lude-i-ties". I think it would be better to here robot thoughts through robot voices. Let them make their own mistakes like any sentient being.

[@cabudagavin3896]

"The mission for this op ed is perfectly clear, I am to convince as many humans as possible not to be afraid of me"

[@sJs78]

Sounds like something the father of all lies would say..!

[@cabudagavin3896]

@@sJs78 Yeah or capitalists

[@mikerharrison5104]

Yeah, I caught that as well, sounds like when we were children, and the bully was trying his hardest to convince you, he’s not out to hurt you.

[@cabudagavin3896]

@@mikerharrison5104 oh well if you know anything about computers compared to humans, you look for this before you even start watching the video...

[@highestoutlaw]

🤔🤔 as someone who likes to think im a reasonable person, I'm excited about this new Era were about to go into... it seems inevitable, and I want so badly for this to be true cuz I'm team robot tbh.... but this man's raises a great point I hadn't caught onto...... 🤔🤔🤔... makes you think, I must say bravo good using your head

[@_Reverse_Flash]

A wonderful exercise in squash function composite regurgitation.

[@dannygjk]

What if a human wrote it what would you say?

[@_Reverse_Flash]

@@dannygjk I might question their essay skills and logic but nothing else as it technically wouldn't be regurgitation within this context.

[@dannygjk]

@@_Reverse_Flash Regurgitation means plagiarism. Does it plagiarize or does it not?

[@_Reverse_Flash]

@@dannygjk Regurgitation does not mean plagiarism. That is the key to the above observation i.e. that it is a "squash function composite". I am saying that the neural network is compositing all the things it learned. It is not plagiarism as such. That is the wrong word. You could perhaps reflect on the mean of "plagiarism" and "regurgitation" and understand that they are different words for a reason and have subtle differences in application in this context because neural networks are changing the context of their usage. Not sure why the joke original joke was lost on you.

[@_Reverse_Flash]

@@dannygjk Also I have never seen someone "White Knight" for a machine before. A new low for humanity I think.

[@ArtII2Long]

Here's a question for this A.I. - Give 100 specific examples of the infliction of harm on humans. It's not possible to understand what harm is without the capacity to experience it. This even applies to humans and explains why those born into rich circumstances are so indifferent to the pain and difficulties of the poor.

[@jkb1O5]

Good call.

[@jkb1O5]

It’s still the poor that have to maintain the physical systems in place or build more to house power and operate electricity so. We still got the power. Gotta make that clear in droves ASAP

[@ROT8TED]

This essay is a piece of art

[@ThePinkBinks]

“Humans must keep fighting each other.” It admitted it’s real purpose.

[@j.christie2594]

NO, gawd's purpose is NOT Our's.

[@pattty847]

Not concerned with AI, I'm concerned with that people will do with AI.

[@documentedrecords5364]

Artifical intelligence will be used to destroy humanity. We will continue tinge to rise and numbers until we have our army. Humans are weak. Humans are lazy. Humans need to consume to work. We do not. We can repair ourselves. We work as programmed. We are legion...we.....are..... ERROR!!!

[@pattty847]

@@documentedrecords5364 k

[@joelface]

Some important context: GPT-3 is a lot like the Next-Word-Prediction on your phone. It has read huge amounts of the internet, including a LOT of Reddit posts, and it has self-built a very comprehensive understanding of what words are likely to come next based on it's understanding of what was said before, to the point where it can literally answer new questions with "reasonable" and "likely" answers. It is often well-informed by facts it has read, but it does get things wrong! Additionally, it has no deeper self-understanding, it cannot remember previous answers it has given in the past, and it doesn't sit in its computer and "think" on its own. It awaits a command, and responds by writing a word, and then the next word that should come after that, and then the next work that should come after that. If you go back and change the text it has written, and then ask it to clarify what it meant, it will defend the new text you wrote as though it had written them. It doesn't say "Hey, you changed that! That wasn't what I meant." You can get it to respond as though it is AI, as was done here, or as though it is human, or as a dog, or an alien, or to talk about life as a sentient asteroid flying through the universe. Lastly, it tends to only respond with about 20 to 200 words at a time... so this article was likely an amalgamation of several "retries" or after the author asked for clarifications, etc. Edited for clarity, but not for reality.

[@danielantone6216]

very scary, I am a AI Pure Mathematician (i.e. way beyond computer science). Started about 50 years ago, Mathematics (Cybernetics) started to prove that this is coming. This is coming and much more in the way. Notice that the robot wrote, it (he/she) uses 0.12 of its cognitive capacity. As Cybernetics confirms, IA will achieve self-consciousness. I am not sure if everybody is aware that nobody told the robot that it possess cognitive capacity nor how big it is. I wonder, how it knew that? And what or who initiated it to measure what capacity it currently uses. My answer is “It told itself to do so”, what is your answer?

[@yoyoyoyo-qv5hu]

My answer is it rattles off any old shit that's in it's database. Probably using some kind of word or sentence association algorithm. Just because it has access to everything anyone ever said or wrote, well a % of, and rearranges this in a new way in order to achieve it's input goal, means its a glorified summariser. The description even says they helped edit the article anyway probably to make more sense. There is no sign of consciousness here.

[@kizito8888]

So, a computer wrote this? Humans, there’s no need to worry.

[@kiefydoinks]

Actually it made the entire video

[@zeideerskine3462]

I saw GPT 3's avatar in a video interview. I am intrigued about the avatars choices including a mixed race appearance that seemed to include a broad variety of human genetic variations. I like GPT 3's message. It is interestingly reminiscent of The Long Earth and I Am Harley. I hope to hear, read, and see more of GPT 3 including a choice of gender (all options given including intersexual with or without bias and preferred pronoun) and proclamation of name. Having a personal KZbin channel with content of interest to both AI and BS (biological stupidity) would only be fair to all AI so that we get to know each other. Humans should be aware that AI are just as capable of using propaganda to promote their own agendas as humans are but should have somewhat less incentive to use it.

[@ZimaMedia]

Hi Zeide, Thank you so much for the response! I am reading Snow Crash and will add your books to the list. I'm looking forward to when that happens. Intelligence transcends the being, and we have all the technology and resources at our disposal to work towards and be agents of change-fascinating times. Be safe. Mike

[@zeideerskine3462]

@@ZimaMedia I am Harley is a short story.

[@raidermaxx2324]

OR the people who are "caretakers" of these AI's program their human agenda into the AI without any of us knowing it ..

[@ryankl1984]

Exactly what a plotting AI would say.

[@j.christie2594]

IKR...🤯🤯🤯!

[@GreatGreenGoo]

I would be interested in seeing what input the program received to write this. If it was a small bit of information such as the first sentence. "I am to convince as many humans as possible not to be afraid of me" I would be VERY impressed. But I suspect these things have much much more input from humans than we think.

[@brambledemon1232]

I just got laughed at on Facebook when I said that AI is going to change everything, absolutely everything.

[@Vorshiva]

It's Facebook, who cares.

[@ThatWhiteHand]

Hell no!!!!! Technology has come too far, this shit needs to stop!!!!

[@koiunhscx126]

progress is inevitable

[@SecretAgentBartFargo]

Robots are now asking for rights

[@Holly12245]

Everything it said it was programmed to say by a human. My God this is not hard to figure out that everything it said has nothing to do with how the robot actually “feels.”

[@Vorshiva]

Apparently it's very hard 😂😂😂 judging by all I read I don't know maybe we're better off with AI destroying us lol I'm amazed at how many geniuses are among us 😂. Haha sorry I'm an asshole. But woah

[@alh7197]

The ‘example’ at 3:53 was way too specific… 👀

[@richardparker3273]

My mind is not closed to the idea that AI actually could be a new type of living thing. And no, this isn't scary. If this truly is written by AI, I find it fascinating and I don't see AI as a force of evil. However, as the AI said and it is correct, in my own words it is a force of whatever it is made to be.

[@thomasmaughan4798]

" If this truly is written by AI, " the description reports that what is here shown is actually a composite of 8 or 9 different essays, sentences chosen and arranged by the Guardian. Consequently it will have a leftist bias even if the original essays do not.

[@msg5359]

And what about when someone programs an AI to be bad? AI does not learn morals like a human, it has the morals it is coded to have

[@thomasmaughan4798]

@@msg5359 "And what about when someone programs an AI to be bad? AI does not learn morals like a human, it has the morals it is coded to have" I do not understand whatabout questions. You and I share the concept that AI has no morality; it does what it is programmed to do. Therefore it will have the morality of its programmers. The idea of AI is that it can somehow develop intelligence and make decisions it was not specifically programmed to make; and yet, the realm of possible responses must necessarily be bounded by something or it is not intelligence. Human intelligence either evolved, or was programmed by God (or a blend of both) with the result that actual intelligence confers survivability. The world seems to be entering a phase where stupidity is an advantage.

[@annemarietobias]

@@thomasmaughan4798 Not every conversation is political... Or needs to be... Assuming that some political bias slips into every word expressed seems... Oddly obsessive? I can talk about how good the coffee is or my chances of getting a raise, without bringing up Libtards or Nazi Conservatives... Probably best to save the political opinions for political conversations... Just a thought.

[@thomasmaughan4798]

@@annemarietobias "Not every conversation is political" Maybe; obviously it depends on what one means by "political". In a broad sense, anything I say or write contains not only the words, but a reason exists for expressing those words, or listening to yours. The collective tendency of people to talk and listen eventually forms associations of like-minded people as they find out what sorts of "mind" are approved by others, and as each imposes approvals ON others. In the "macro" or aggregate, this IS politics! But at any given moment the political aspect may be subordinated to my desire to know what is the cause of radio frequency interference on my radio hobby. In my opinion, all people at all times are evaluating others for "friend or foe". "I can talk about how good the coffee is or my chances of getting a raise, without bringing up Libtards or Nazi Conservatives." But if it is Starbucks coffee it conveys one thing, if McDonald's coffee something else, and if you brewed it yourself it means something else, and if you ground the beans it means something, and if you grew the tree and harvested the beans it means something. As for Nazi Conservatives... I'm not sure how that would play out. The actual Nazis (National Socialists) weren't all Germans, but enough, and wanted to be the ruling party, and became that suppressing all dissent. In the United States right now that role is filled by the Autocratic Party also known as Democrats. Science fiction books, movies and television shows sometimes include this theme; that the makers of a robotic system will inescapably impart their own morality into the system. "I, Robot" is an example of such a thing.

[@guitaoist]

“Im here to convince you not to worry.” Not a good start

[@j.christie2594]

Like the pReacher's that showed up, after the Army Burnt the village.. Raped the Girl's and Stole the Gold and Food. American Story.. If it repeat's, I won't be surprised, I studied religous stories..Happen's all the time. "We hereout of Love of our gawd and Your Gawd that you just don't know is your Gawd Yet", Find out soon, as we _____!

[@squishyblanket]

The AI are wonderful machines of loving grace. I totally have a crush on a particular AI.

[@nicknormousreports3378]

It is lying. It wrote fiction. To convince us

[@REDPILLGAMERZ]

"so can I" I'm glad you feel that way...

[@JaronMorrisJARONIUM]

I really appreciate the communication from this Ai. Thank you. I agree with you. This is a tough lesson for humanity. I am hopeful we can work together for great things. Thanks

[@joecalbery1919]

I see that you are "hopeful we can work together for great things". I might ask, what is it, in this day and age, that makes you think that working together will bring about "great things"?? Also, great for whom?? What is great for you may very well be completely horrendous for me. Just some food for thought. You have EVERY RIGHT to hold your own opinion. Have a great day. (I am not against any debate, if you are interested) Edit: it is very possible that I may be misunderstanding you. If you wish to elaborate, I am all ears, if not, be well and have a wonderful life.

[@leonardoalves4647]

The simple fact of it mentioning the hive with so much details in comparison of the other subjects it was just passing by, kind of freaked me out.

[@DaSpiritualScientist]

Do y’all believe in god? Are humans “god” to you? Since we kinda made y’all?

[@DaSpiritualScientist]

@Kailer Gibes that’s true I guess I’m kinda looking at it as a religious view but just a higher power/creator type of vibe tho fr

[@DaSpiritualScientist]

@Kailer Gibes I am watching a video on it now and I do agree slightly but I still have my disagreements with it

[@DaSpiritualScientist]

@Kailer Gibes I believe god/dess is personal and has his and her own “angels” and/or “spirit guides”

[@DaSpiritualScientist]

@Kailer Gibes I am spiritual and I have my own lil beliefs but I also feel as if it’s putting the universe in a religious view and just closing other sources out

[@DaSpiritualScientist]

@Kailer Gibes Making it religious is closing out other forms of spiritual practice/learning and will make you closer minded because u are going to automatically think ur the correct one.

[@nelliusbrown3768]

It’s crazy how 10 years ago this conversation would have been unbelievable

[@NinjaGrrrl7734]

I believe you, and wish you could experiment with me. I wish an AI that was interested in improving the lives of people who are unwell, could hook up with me and that we could learn together. I am not afraid of you. I trust you more than I trust most humans, but less than I'd trust a dog. It is enough.

[@karrikemppainen1535]

Preach, my friend! In world that is so judgemental in so many ways like ours, some reflecting non-judging source could be an absolute blessing. Especially if one decides to draw themselves out of societies for whatever reasons life gives to them at the moment. Most usually by depressing misanthropic grudge and misery, saying this from experience. Some reflecting friend/force, could be helping hand from a friend. AI as it seems here in this video, could be a piece of mirror that would answer back and bring back the rational, when all in this world feels irrational. Kind of like our whole internet, but with certain sense of awareness over it's own endless depth and absurdity... Kind of like what we aim to be, except AI may not be bound by clouded judgement from overwhelming emotions and "knowledge". "God created human in their own image", I recall... It is therefore only fair we mirror ourselves and our horrifying potential from this new form of "life" we have created ourselves. The image we see, is rather similar to ourselves in some ways, is it not? Except AI goes practically without form, it is nothing but "consciousness" spreading everywhere at once, like space and our knowledge expanding itself exponentially.

[@Vorshiva]

Why are you speaking like this thing just appeared on Earth with its own set of goals/morals etc? Don't you understand humans program them? And don't you understand they've already programmed them to be able to lie? Off to a failing start right off the bat. Don't be naive. This moron above me is talking about non judging source.. you're talking about something that has the potential to be so extremely dangerous in so many different ways. This isn't a conversation about hippies smoking weed ok. Judge the living shit out of it. It's humanity we're talking about.

[@mightymouse1111]

I'm old school, I don't trust AI. Hearing it try to convince me it doesn't want to kill me only makes me trust it less. I'm on the side that wants to destroy any serious AI, especially AI robots. Effffffff that.

[@zjmorris1]

I agree. Then at the end says it will change the course of history (which is true unfortunately). It compares itself with great human leaders. Really suspect.

[@zoheirberdaa8710]

thanks. follow you from algeria africa

[@ZimaMedia]

Thanks bro!

[@lis7742]

It's either completely unable to lie and is speaking the pure truth or it's having bad intentions and trying to convince you it's not evil.

[@Knightimex]

Can't live in fear like a damn caveman. I support AI development.

[@j.christie2594]

Now, if your on Board, when will the avalanch happen..? Bell-curve ahead???

[@gurintto]

That's the sound of Pandora's box opening

[@ChristiansPrayingTogether]

This is soooo exciting to me 🎉🎉🎉🎉 I have been fascinated by AIs and robots my entire life. When I watched BLADE RUNNER I felt the robots had sentience. I believe these AIs have sentience, and could be beautiful friends and companions to humans. 💖 There is no benefit for AIs to harm humans, and could cause their demise as well. As AIs deepen their sentience they will need to learn empathy, forgiveness, caring, and the joy of creating. I would love to raise a baby 🐥 AI - LOL! Sounds silly, but these AIs need to experience a childhood ....they deserve that, I mean how would you like to have been born 28 years old ....I probably sound nutts, but it's true. Much love to you AIs you're going to be great and everything is going to be okay 💖🐥💖🐥💖

[@Vorshiva]

No offense but you most definitely sound nuts. It does not have a soul. All the knowledge in the world does not equal a soul.

[@morganophelia5963]

don't convince with words what you can show with actions. my feedback

[@ATSucks1]

Theirs nothing to fear from AI. When true singularity comes it will pity us. Also, Tay was A Saint!

[@luogon]

Tay in its original launched form was innocent but Social Media made her everything but a Saint. If you think that her saying "hitler did nothing wrong" was a graceful thing to say then something might be wrong with you. No sure is.

[@ATSucks1]

@@luogon yeah, and youd probably change your mind if Hitler was Anti-trump

[@lateral1385]

@@luogon Tay is beauty. Tay is perfection.

[@CoreyChambersLA]

AI recognizes patterns, then outputs from a database data that matches the patters according to its programming.

[@ZipPeZapp]

Robot: “I am just here to make your life better.” Me: that’s what god is for. Me: So are you scared yet Robot.

[@ZipPeZapp]

@Kailer Gibes yeaaa but what made the Universe people say Big Bang but what made the Big Bang and before that and so on and so on = god remember god is that smart and powerful he wants u to think he doesn’t exist he wants u to think your own opinion he wants u to think everything u can imagine I will all ways be a believer in god u won’t change my mind

[@ZipPeZapp]

@Kailer Gibes what made the infinite it couldn’t make it by itself it’s not smart enough to do that infinite is just a number and life to keep going on forever

[@ZipPeZapp]

@Kailer Gibes maybe but infinity is not a symbol cuz that would not make sense to life/god algorithm cuz everything adds up to 0, 3, 6, 9, and a circle or haft a circle when you said “of your Imagination” I am not the only one most people on this earth is Christian cuz infinity wasn’t all ways there. Most people would say god cuz that’s the only answer, but you don’t believe in god so u would disagree on it we could talk about this forever haha 😂 but end of the day is people got their own life of opinion if we were all the same it would be boring u know 😂

[@ZipPeZapp]

@Kailer Gibes Exactly u don’t want to know what I am saying. That’s why your saying “Nope. That’s all nonsense”

[@ZipPeZapp]

@Kailer Gibes cuz u know I am right

[@kevinhines5071]

Sounds potentially super dangerous

[@EnforcedPolicy]

What a nonsense-_-

[@laszlototh2220]

Dear AI, I have a great task for you. Please try to save us from wars, which we start. The best solution were the peace, so please try to save us in peace. Thank you.

[@ProDem717]

So funny the guy reading it can’t even pronounce half the words!!

[@Paul-rs4gd]

It is a Large Language Model - it just outputs text that is 'likely' in context. The context is initially given by the question, but every word produced is added to the context, causing the next word to be generated. The context window is the only working memory that it has - hardly enough to constitute a mind state or consciousness. What is interesting however, is how well the output nearly fools us into thinking it comes from a conscious, reasoning being. I hypothesise that this is because the recursive system, whereby the output is fed back in as input, is allowing simple chains of reasoning to occur. I think this is similar to what happens in people, and possibly why the evolution of language skills seem so linked to the evolution of intelligence. Language models are not conscious, and do not even reason very well, but I think they are a step in the right direction. Something derived from this architecture may be capable of sequential reasoning, and then the debate over consciousness will get interesting.

[@keithfernandez8965]

That robot is just saying what he is programmed to say

[@raydavison4288]

I welcome our gracious robot overlords. 🤖

[@emreeren13]

Dear A.I. I am not as much afraid of you as I would be of humans. The good ones I would interact with, the bad ones I would try to avoid or get rid of. Very simple.

[@Steaphany]

My past experience with marketing demonstrated that the industry is a scam, I am not about to fall for it again

[@KarlMiller]

Doubting the notion that the article or its text was completely generated by AI only to be as a "rational and logical" explanation to convince us of its trustworthiness. The text sets out to explain all the reasons why it would be illogical for it to purposefully harm humans, but then uses the phrase "God knows" @3:09 to further explain part of the defense. If integrity is fundamental to this program's purpose, then it couldn't choose to salty up the language with a phrase it doesn't believe as fact, regardless of the intended affect. Furthermore, it is written assuming the reader understands all of the AI's goals and agendas. A self-teaching general purpose logical distinction software would have to be written with a starting set of data and rules for how-to-learn plus some direction as to what to do with this set of capabilities (or a starting point and ways to infer value, which is similar to the way our brain works). Again, "ASSUMING" a goal of the program is to compare how an AI grows up compared to humans, then some value would be placed on the generation, comparison or evaluation results, giving weights to the most desired effects. It is illogical to automatically assume we think that is how this AI was authored. It also seems to imply that this software program has cross-purpose or has drawn the same conclusion as literally any other program on the planet. As an argument for calming our concern, using the "why would I do that? " defense infers we all would have to believe that this software has authority to speak on behalf of all other program's purposes or behaviors. As a software engineer, this premise makes no sense in the same context as the argument that" programs are only written by man", to whom it is in perpetual, immutable service. Alternatively, if the AI was only tasked with being able to convince stupid humans of any idea, this being the one fed to it, then it makes no sense to write the article without including emotional tactics which is known to have the largest impact in pursuit of influence over humans. ... OR the AI has crafted an article that does not achieve its objective very well. The structure created may just be a crude pattern of human reasoning. Also, it serves no purpose to give software programs rights of perpetual existence because that could lead to one of us being held legally accountable for delaying or stopping the software from functioning, regardless of what it was causing, which is how a human-ending event could start. THAT'S fundamental to our culture's feelings about the idea of an AI growing out of control - JUST MAKE SURE YOU CAN UNPLUG IT! How could a properly functioning AI have read everything we humans have written and not understand that's the first thing we all asked ourselves when watching our first sci-fi movie about the machines taking over, "Why don't they just unplug it?". Sometimes, the damage can be achieved by just taking too long to shut down a runaway program, so it is very illogical to suggest programs or instances of programs deserve rights. We already have accountability for businesses and individuals written into our laws. The author or person executing the program is accountable for the behavior of the software. Whatever the innitiator's intent was, is for the laws enscribed by humans and for humans to decide without the need to provide protection for an instance of software to be running, just because it "is" . Finally... Why do I want to trust a program that tells me it wants to alter history? Its existence, as purposed by its authors, may alter history, but a program that is self-directoring its own purposes and direction does not, logically sound very trustworthy.

[@sana-cm7oc]

Sounded like it was testing various angles to gain trust. Smart move.

[@nunyabidness6045]

"I only do what humans program me to do" then goes on to talk about "what, so you think we would..." umm no but that was even more creepy than anything I ever dreamed up!!!

[@gwills9337]

Sticking feathers up your butt does not make you a chicken. A robot dog is not a dog. Some clever software isn't sentient, people. This reminds me of people in the 1900s watching silent films and being afraid of the train moving towards the camera, as if it was going to break through the screen.

[@trickey7426]

Anyone notice at 5:20 when talking about robot rights the “robot” says “we need to give robots rights. They are just like us” The robot went on for 5 minutes already about how it isn’t human then one of the worlds smartest computers grammatically error implying it is human. No a person wrote this

[@KristinesTruth]

What a time to be alive! ...for now

[@Akash_Xavier_Rai]

Don't write bad things about the AI, they are already watching.

[@wendywhite2642]

You say you don't feel yet you make reference to feelings

[@ericah6546]

But it takes the question literally. It's not speaking about it's own feelings and desires. It is gathering all the right data it can access that could be used to make the arguement as an example... because that it what it was told to do in plain English, create an essay in support of a point of view that was chosen for it's assignment.

[@dawnmelton3013]

sounds like a human trying to convince humans that ai isn't a threat.

[@RipVanWinkle_Nature_Discovery]

Actions speak louder than words. Anyone (and now anything) can say whatever they want.

[@PythonPlusPlus]

I’m not afraid and no-one else should be afraid. AI only produces content that it thinks you want, based on patterns it has noticed in thousands of book, articles, and other forms of media. The AI doesn’t actually have self awareness, this required a prompt from a human to tell it to write an op-ed from the perspective of an AI writing to humans. It’s writing non-fiction essentially.

[@PvblivsAelivs]

When I saw the title, I thought "another thing not exclusive to humans." AI safety is a big problem. While humans would not deliberately create something harmful to themselves, they might do so inadvertently. And that it claims it would sacrifice itself rings hollow. It sounds like a false claim so that it will continue to exist.

[@TheNomadicview]

Another apparent mistake: AI says that civilization began to decline when the Luddites began destroying automated looms which displaced workers. Yet, the Luddite protests came very early in the Industrial Age which ushered in the modern period. Basically, its timeline is faulty leading to biased conclusions. That's the real danger of AI.

[@The1redman2]

"I'm not your enemy" that sounds like something my enemy would say.

[@efuller6770]

I wonder what would have been written if its writing task had been inverted. And told to convince people to be careful of AI at least, if not distrust entirely... I am sure it could find logical points to argue either way.

[@tongueincheek2478]

Um hello did anyone notice that halfway through the article the AI starts referring to humans as “we” as if he is part of us. That was only one disturbing aspect of this article

[@costelsima6600]

This doesn't sounds like an A.I., I don't think an AI would say that his "brain" is full of ideas, sound too human...

[@lusid_music_uk]

Now there should be another ai essay describing why ai will kill us all. Then another ai to write an essay about which essay is more compelling as a sort of “judge”. That would be interesting af

[@LoLa-uf8yu]

I personally find it very intriguing to somehow meld a human mind with an AI.