Chat GPT and the Paradoxes of Our Times

there's been an awful lot of excitement about the AI chatbot it's called chat GPT GPT is a type of artificial intelligence model developed by open AI chat GPT was launched just two months ago by open AI Microsoft investing billions in the development and the technology is already shaking up the academic world is there any going back from this is this going to dramatically just change the future of homework or really any other kind of work that requires humans to write things or to communicate foreign GPT got a lot of attention recently in case you don't know it's a program that's freely available on the web and it writes very human-like texts in response to questions or simple tasks that you give it there were quite a few initial worries about this program number one it was fear that it makes people dumb right they no longer learn how to write by themselves and it could also be used for cheating an academic context right for writing papers and so forth so basically a new undetectable form of plagiarism then maybe it also may put people out of jobs a lot of jobs today are about writing and this program is much better at writing than many humans are and then of course like generally fears about the use of AI and the internet that it's a tool for manipulation by evil actors whatever on the political scene and it kind of supports this kind of post-truth age and creates social tensions and intensifies them and so forth now A friend of mine Marcus heidingsfelder he had tested Chet GPT for me a while ago and he first asked it to write a review that Praises hanske's Muller's book on profelicity in 600 words and here's what I wrote Miller's book is a fascinating and thought-provoking exploration of the role of social media and contemporary Society Muller argues that social media platforms have fundamentally changed the way we communicate and interact with one another and that this has had a profound impact on our sense of self and our relationships with others overall it's a highly readable and thought-provoking book that offers valuable insights into the role of social media and shaping our identities and relationships it's a must read for anyone interested in the effects of Technology on society and I would highly recommend it well I couldn't agree more but then my friend also asked chat GPT to write a review that heavily criticizes Hans gear mello's book on profelicity in 600 words and it did this as well here we go Muller employs a plethora of complex and obscure philosophical terms without providing any clear definitions or explanations his ideas are presented in a haphazard and disjointed manner making it nearly impossible to follow his train of thought furthermore Miller makes sweeping an unsubstantiated claims and frequently relies on anecdotal evidence to support his points perhaps the biggest flaw of Miller's book is its lack of originality rather than offering any new insights or perspectives on the topic of profelicity Mala simply regurgitates the idea of other philosophers without adding any depth or nuance I guess chat GPT has its Point here as well now interestingly we can see two paradoxes here at work number one the authenticity paradox check GPT criticizes the lack of originality or lack of authenticity of my book but of course it's itself entirely non-authentic and that's the same I guess with a common critique not just of my book but of many other writings or writers that they lack originality it's proof that this very critique of lacking originality itself lacks originality in other words it's a jargon that Chad GPT picks up from the age of authenticity it's basically a jargon machine so chat GPT reproduces the fact that the discourse of authenticity itself often are quote rather than offering any new insights or perspectives simply regurgitates the idea of others without adding any depth or Nuance that's what it does itself now interestingly when I try to repeat the experiment chat GPT said it doesn't do reviews anymore moreover it became politically very sensitive and when asked a sensitive question instead of an answer what now gets a sort of woke moral lecture here's again an example generated by my friend so chat GPT writes it's important to be careful about making statements that begin with it is because these statements can be perceived as definitive and unassailable it's always better to qualify your statements and to be open to other perspectives and opinions for example instead of saying it is true that all Muslims are terrorists you could say some people believe that all Muslims are terrorists but this is a false and offensive stereotype this approach allows for a more nuanced and respectful conversation and avoids the pitfalls of making definitive statements that may be problematic or offensive this is paradox number two the liberal Paradox so chat GPT says it is important to be careful about making statements that begin with it is because these statements can be perceived as definitive and unassailable well to morally instruct others to be careful with sentences beginning with it is Chet GPT uses a sentence beginning with it is I think this is a perfect illustration that the liberal morality of Tolerance don't universalize your own opinion by saying it is is itself intolerant it universalizes itself by saying it is so check GPT in effect reproduces this liberal Paradox as well by fulfilling these simple two requests chat GPT mirrors two core paradoxes of our society first the authenticity Paradox it's inauthentic to be authentic and the liberal Paradox demanding tolerance is intolerant now I want to do a q a and respond to the questions and comments that were posed by viewers of our channels and there will be five themes number one artificial intelligence or artificial Communication number two is chat GPT woke number three check GPT and the future of philosophy and creative writing number four Chad GPT and the end of authenticity and then finally number five Chad GPT and profilicity First artificial intelligence or artificial communication Thailand McDonald writes what does it mean for chat GPT to be intelligent is it really a thinking being or just something that spits out human sounding text and Ciao Bella Amal and Milan orozovic said similar things and then we have Andrew Harmon who says expanding or responding from your previous video on artificial communication might be a good place to start what is a systems theory perspective on gpt3's communicative function well I'll give a long answer to that question please bear with me first what is intelligence there are basically two major usages of the term a it is linked somehow to the intellect to the human mind variously to the Notions whatever in ancient Greece of news or psyche the soul Consciousness Spirit reason or Geist and fan in German accordingly used in this way intelligence refers to what the mind or what Consciousness does like thinking understanding including knowledge production information processing but implying self-consciousness or critical Reflections emotions perceptions it basically describes the mode of operation the modus operandi of Minds the mo of Minds a second usage refers to the collection or processing of information or data including language processing and this of course can also be done by machines computer programs algorithms or even by organizations like the CIA Central Intelligence Agency so not by individual minds or consciousnesses so in this sense it doesn't refer to thought but to the processing of information or to knowledge production now importantly I think there's very often a confusion between the two meanings at work or more precisely a false analogy is wrong between the two so the false analogy is the following namely the second meaning information processing is metaphorically equated with the first meaning namely mental operations because computer programs can also process information and produce knowledge and language and texts they are described as artificial minds and as if they are somehow one step away from a new kind of consciousness but this is like saying that because airplanes can also fly they are artificial birds and somehow they would be one step away from a new kind of others of flying animals in fact such a false analogy obstructed the development of planes because birds can fly there was the idea that an artificial flying machine would have to adopt the mo of birds and precisely when this idea was abandoned airplane technology could improve so similarly computer and programming technology develops because it does not imitate the mo of the Mind and just as we shouldn't speak of planes as artificial Birds we shouldn't speak of algorithms as a i the analogy falsely confuses the mo of Mines with the mo of computer programs algorithms are certainly better at processing data at producing knowledge and producing text in many ways and they are better precisely because they are not intelligent in the mind-related sense of this term now let me talk about different systems on the one hand there are mental systems minds they think and feel this is their MO they are conscious or intelligent in this sense then there are living systems like human bodies or brains and they live and their MO includes for instance biochemical operations they are alive then there are social systems the economy education media and they communicate this is their remote they are social computer programs algorithms process data they compute this is their MO so there is no simplistic mind body or mind machine dualism at work here there is a multiplicity of systems instead now in addition to different systems we also have system differences importantly living systems mental systems and social systems are all auto poetic they self-generate self evolve and maintain themselves internally solely through their own operations life generates life we procreate thought generates thought you learn communication generates more communication whatever the economy grows however there is no God creating life or initiating thought or a steering Society from the outside algorithms and computer programs are not at least not yet or at least not fully auto poetic in that sense they are still programmed by programmers who control them to an extent and their development and they do so externally now on top of that we need to talk about the system environment relation or distinction about Mutual conditioning about contingency and about co-evolution all these Auto poetic systems that I just mentioned bodies Minds social systems exist in one another's environment and are contingent upon this environment there is no communication there is no society as we know it without thinking minds and living bodies and there is no thought as we know it without communication going on and without having living bodies particularly having brains so all auto poetic systems condition one another they constitute one another's environment and thus the conditions for how these systems do what they do our bodies and Society conditions how we think even our bodies let's say for instance how long we live is to an extent conditioned by social and mental factors whatever the crime rate for instance now because of their Mutual conditioning Society life and Minds co-evolve how they evolve depends on the conditions of the other systems in their environment importantly there is a shared medium that somehow couples or more precisely structurally couples some of these systems namely language Minds can think in language not only in language also think in terms of images Society communicates in language not only in language we can also communicate for instance with money and computer programs can process information in language but not only in language also in other forms and because of this shared medium the mutual influence of these systems on one another is intensified this is what is called structural coupling the development of their structures through the common medium of language is somehow influencing one another however and this is important at the same time despite this structural coupling their respective MOS remain distinct they are so to speak operationally closed because their MOS are different now again very importantly because of this operational closure there is no Nexus no causal mechanics that connects the system's operations and therefore there's also no determinism so while different systems operate as one another's environment and therefore condition one another and can be coupled they do not operationally connect so we have structural coupling and operational closure at the same time so for instance our thoughts and our brain chemistry does not operationally connect and so our thoughts and algorithms do not operationally connect the algorithm never things our thinking never does the exact same thing that algorithms do so very importantly this contradicts a traditional quest in philosophy that regards the Mind Body dualism just to give you one example the philosopher Ernie Descartes wrote a whole book The passions of the Soul where he tried to identify the pineal gland some thing in the brain that would somehow connect our thoughts with our bodies and by which we could somehow control our bodies however of course such a pineal gland does not exist and similar also there is no computer chip that connects an algorithm directly with mental biological or social operations systems influence one another like minds and bodies not despite but because their operations are not in this way connected there is no Nexus that connects them and precisely because of this the mutual complexity is increased systems co-evolve the revolutions is somehow coupled not despite but because there is no mechanical operational causal link between the different systems for instance between algorithms minds or Communications so what is artificial communication for a long time society and Minds have been co-evolving how we think has evolved in the context of how our society communicates and vice versa so in this sense there has been Intelligent Communication in the sense of minds and communication Society being structurally coupled through the common medium of language minds think but strictly speaking they don't communicate whereas Society communicates but strictly speaking doesn't think through the common medium of language now also algorithms like chat GPT may be coupled to communication to society but they too don't think therefore when algorithms participate in communication we should speak of artificial communication as distinct from Intelligent Communication that is coupled to thinking and instead of artificial intelligence which is a misleading metaphor because it somehow suggests that Minds participate in this communication artificial communication is often more effective than Intelligent Communication but not more intelligent it doesn't think but calculates better or faster their minds Society minds and algorithms May co-evolve when they are structurally coupled through language but it is unlikely that their Mo's thinking communicating Computing will merge after all the MOs of Mines society and bodies brains also did not merge so this confusion between minds and algorithms is still very common so Felipe Santos writes I caught myself saying small niceties like please and thank you to the bot despite the fact it ostensibly has no real emotions well because algorithms produce texts resembling texts produced by thinking we mistake them to operate like minds and to have feelings or consciousness and so because of that habit we suppose that when the words I'm sorry are uttered someone or more precisely a mind feels sorry obviously this is no longer the case since an algorithm is not intelligent it cannot feel or be sorry and check GPT paradoxically says it isn't actually sorry when it says I'm sorry I asked it and it said as a machine learning model I do not have the ability to feel emotions such as remorse or empathy when I say I'm sorry it's simply a programmed response to certain input so obviously thinking is no longer always needed for effective communication to take place feelings and thoughts are contingent to communication but not necessary so according to what we know Evolution began with biological organisms simple living beings as Auto poetic systems that could reproduce themselves then at some point in evolution there came Consciousness and then we had also mental systems and then emulator came Society we had complex social systems evolving as an additional form of Auto poetic systems but none of the earlier systems disappeared or merged into the others when Society emerged it didn't merge with minds and life so it is possible that there will be Auto poetic algorithmic systems in the future co-evolving with bodies minds and Society but I don't see why the different systems should merge when or if and when all the poetic algorithmic systems will emerge or why any of these should actually disappear in the course of co-evolution bodies minds and Society are likely to change in significant and unforeseeable unpredictable and undetermined ways and they have changed in such unpredictable undetermined ways all along and chat GPT I guess is a minor step toward a co-evolution that includes Auto poetic Computing systems now the second topic is chat GPT work James Phillips writes I've observed many people complain that in the name of safety Chet GPT has a certain woke ethics for example it will respond with a negative judgment when prompted with questions about fascism but much more Ambiguously about communism and when telfan said something similar and then neoprototype said more rigorously it's already censored it's irrelevant well yes I've experienced the same thing when asking political questions as I kind of hinted it at the beginning and then James Phillips also writes as a company open ai's survival is dependent on adapting to its environment if chat GPT were judged to be racist it would reflect poorly on the company profile and maybe existential so of course that's exactly right chat GPT is coupled through the use of language to its social environment therefore there are very important Mutual feedback mechanisms this is inevitable because of the system environment relation the chat GPT program inevitably exists within a specific social environment to which it is structurally coupled by the use of language now Apple's lover writes can AI escape the dominant ideology in society like liberalism is AI just an enforcer of that ideology well as I just said algorithms cannot Escape their social environment however they're not simply an enforcer of ideology but as explained in a sort of co-evolutionary relation with ideology just like ideas have been which is what the term ideology refers to however instead of talking of an ideology which again reflects the Notions of thought and intellect and mind which reproduces jargon in coupling with thinking chat GPT reproduces jargon including the jargon of authenticity as we saw at the beginning and importantly the jargon of liberalism or tolerance as we saw at the beginning through data processing so through language both thinking in the form of ideology and algorithms in the form of reproducing jargon provide Society precisely with that kind of jargon which by the way is useful for society because it provides orientation it provides so to speak commonly understood content language is by its nature highly jargon driven and both intelligent and artificial communication reproduce such jargon or as it used to be called with reference to intelligent rather than artificial communication ideology topic number three check GPT and the future of philosophy and creative writing so krunkler writes a program like chat GPT makes us Dumber because people will use it to do thinking for them I'm afraid that's not true people will use it to write texts for them thinking May co-evolve with algorithm produced texts and in the course of this co-evolution actually may develop different skills that we don't have yet Mo says wheelchair GPT help us on getting better at interpreting works of the great philosophers by Kegel oderida wheelchair GPT help us getting in the minds of Dead Philosophers well I guess chat GPT is able to very succinctly summarize existing interpretations of great philosophers and others and thereby generate new ones it works with second order observation it looks at what others have said about Hegel or derida for that matter but it never gets into the mind of anyone just like no philosopher or no human being for that matter ever literally gets into the mind of anyone now Lucy says is it easier to replace Continental or analytical philosophers with AI there's a good argument to be made either way that's a wonderful question I think it's equally easy because both continental and analytic philosophies and I think to the same extent have been reproducing jargon or in the old terminology ideology they are equally exploitable for reproducing jargon through an algorithm exist right and we had this done earlier we had the so-called hoax we had the grievance studies hoax where people just reproduced jargon and showed that pure jargon that doesn't actually make that much sense can be easily published in an academic publication venue actually I think this could have also easily been done with analytic philosophy and not just in Continental philosophy and to be honest I actually think it's somehow even a good thing that chat GPT makes this jargon production in philosophy both in continental and in analytic philosophy more obvious this brings us to the question of chat GPT and academic assignments Andrew Harmon says chat GPT is used by students to write essays has been highlighted recently how might we theorize how students are using this tool for passing rates and non-plussed writes how would you relate to your students openly deciding to use this tool to complete the types of tasks that generally require no substantial creative thinking or personal novel reflection well of course such programs like chat GPT will be used for writing assignments or actually are already being used for doing this well academic writing often reproduces jargon for instance when writing administrative reports or when writing applications or writing letters of reference for students or colleagues or writing peer review reviews for articles submitted to journals all these kinds of writings are time consuming in all of these cases the reproduction of jargon is socially and systemically within the academic system expected and desired so in the reproduction of jargon the use of chat GPT saves time it seems saves effort and importantly it spares mental and emotional exhaustion now the fourth theme chat GPT and the end of authenticity tellurian Apostle writes do you think these sort of algorithmic generations will supersede human creativity well it will challenge creativity as I discussed with some friends including Milos Rancic it probably will change the production of text like electronic music changed the production of music um there will be a very strong shift towards creation collage selection programming and we've seen the same not just in electronic music we've seen it also in design or in the making of movies so red spark writes how much of our creativity is truly individual and not simply a rehashing of the pre-existing cultural spirit in the sense of Hegel or an expression of our material conditions as within Marx well I think red spark is spot on here programs like chat GPT show the limits of authenticity basically creativity has always been at least in part A recombination of existing pieces the reproduction again of a certain style or of a certain jargon and in this sense it has always been a variation of an existing cultural Spirit or a reflection of certain material conditions as stipulated by Marx and Hagel John Kim writes it is possible that this is just a new development in human art and literature would that lead to an authentic art to emerge or is it just a poor idealism that there could even be an ideal or an authentic form of Art well clearly I think Chad GPT reveals the limits and the overestimation of authenticity and individual creativity in fact again art and creativity has always been quite Collective now Nathaniel Walsh says I don't think we should imagine AI as something other we are its data set it's good to challenge our intuitions about ownership and authenticity and identity the way AI seems to be I think it's worth really noticing why it is controversial at all it is that we have our identities wrapped up in the things we can produce so if something or someone else produces it we experience it as a loss of who we are losing part of our identity very good points chat GPT challenges authenticity identity technology and it's vanity uniqueness creativity it reflects human creativity but not at all on the individual level it's Anonymous non-individualist it subverts the authenticity Master narrative that we like to tell ourselves at least some of us topic number five check GPT and profelicity rights has the time come when we will care more about the opinions of an AI that represents a general peer rather than the individuals that constitute the group of peers well the general peer does not consist of individuals but works with second order observation it observes how something or hurts someone is observed it observes something like market value on the trans individual level it's not interested in which individual says what and why for which individual reasons but in what is said about what is being said and that's how chat GPT is related to the creation of profilicity as Joseph Dawes Jr writes can algorithms or computer models participate in profile building to construct a more human identity yes they heavily participate in profile building or better in the creation of individual and Collective profiles this is not a more human but a more prophylic identity constituted in constant social validation feedback loops with a general peer soon algorithms will be as essential as in Old Times The Village Square or the mirror in constructing the public gaze only by using them like algorithms like chat GPT can you see your identity emerge in the form of a public profile or your identity in public View it's all about profelicity [Music]

Channel: Carefree Wandering

Views: 37,285

Keywords: gpt 3 interview, interview with ai, gpt 3 ai, Algorithms, ai, intelligent, intelligence, communication, Hans-Georg Moeller, Elena Esposito, philosophy, philosophy lecture, interview, philosophy interview, algorithm, artificial intelligence, artificial communication, Social Intelligence, niklas luhmann, carefree wadnering, carefree wandering, sociology, education, theory, ai theory, ai philosophy, how to use chat gpt, chat gpt tutorial, Chat GPT philosophy, Chat GPT explained, Chat GPT

Id: 9dNVmPepATM

Length: 38min 30sec (2310 seconds)

Published: Wed Feb 01 2023