Generated AI image shouldn't be in social media platforms at all

by WarriorForum.com Administrator
25 replies
Social media is supposed to be platform where content is user-generated. So, what's the point of having user-generated content if all the images we see is AI generated?

It doesn't feel authentic at all.
#generated #image #media #platforms #social
Avatar of Unregistered
  • Profile picture of the author Monetize
    Originally Posted by WarriorForum.com View Post

    Social media is supposed to be platform where content is user-generated. So, what's the point of having user-generated content if all the images we see is AI generated?

    It doesn't feel authentic at all.

    AI images ARE user generated.

    The user instructs the AI image maker about the
    details on the type of image they want, and then the
    AI system produces it. AI doesn't generate images
    on its own.

    If I need an image of a house, it's a whole lot easier
    to develop an AI image to my specifications, than to
    search for it all over the internet, or ride around town
    looking for the ideal house to take a picture of.

    On top of that, when I generate an AI image, I am the
    copyright holder. That means I own it and I don't have
    to ask other people's permission to use it.

    Whether you agree with me or not, there's not much
    you can do to change the fact that AI images are
    going to be EVERYWHERE.
    {{ DiscussionBoard.errors[11791846].message }}
  • Profile picture of the author socialentry
    Originally Posted by WarriorForum.com View Post

    Social media is supposed to be platform where content is user-generated. So, what's the point of having user-generated content if all the images we see is AI generated?

    It doesn't feel authentic at all.

    how is copy authentic?


    its effing hard to find anything with google today.

    take any topics:its almost always some "10 best tips to ... " lifted from copywriter's playbook and the 10 tips are almost always beginner tips or some platitude like "ur destiny is to accomplish ur destiny".
    {{ DiscussionBoard.errors[11791859].message }}
    • Profile picture of the author Odahh
      Originally Posted by socialentry View Post

      how is copy authentic?


      its effing hard to find anything with google today.

      take any topics:its almost always some "10 best tips to ... " lifted from copywriter's playbook and the 10 tips are almost always beginner tips or some platitude like "ur destiny is to accomplish ur destiny".
      Everything it broken today it's almost as if every industry is competing to give a worse customer experience. Cutting costs to increase profits.

      AI will probably quickly hit a death spiral as more and more of the content available to train llm is content generated by AI.
      {{ DiscussionBoard.errors[11791869].message }}
  • @WarriorForum.com,

    Originally Posted by WarriorForum.com View Post

    [SNIP]
    Generated AI image shouldn't be in social media platforms at all
    [/SNIP]
    Really?! A total ban of AI-generated images across the world's most used social media platforms?
    Your proposed solution is something that can only be attained within a reasonable time frame through the concerted effort and mutual agreement of governments, grassroots communities and industries of the free world?

    Why not start with something that can be attained organically through the natural movement of these governments, grassroots communities, and industries?
    I mean, any image, AI or human generated, won't have any meaningfully measurable effect if it can't:

    ** Pass the concerted policing efforts of governments, grassroots communities and industries;
    ** Grab the attention of the publisher's target viewers;
    ** Entice 'em target readers to check out the ongoing relevant conversation;
    ** Make those leads trust the authority and knowledge of the publisher; and
    ** Compel 'em target customers to take immediate action that's mutually beneficial for themselves and the publisher.

    Any content, AI or human generated, still need to qualify for the target audience's current consensus of how they calibrate value.
    So this organically segregates crappy content from that with value. And this takes time. Though not as much time as you'd likely need if you're waiting for a ban to happen in the foreseeable future.
    Signature
    • Deep Learning & Machine Vision Engineer: ARIA Research (Sydney, AU)
    • Founder: Grayscale (Manila, PH) & SEO Campaign Manager: Kiteworks, Inc. (SF, US)
    {{ DiscussionBoard.errors[11791905].message }}
  • Profile picture of the author savidge4
    "Generated" is Fair Use spun every which way... BUT Fair Use is not a bad thing... to better understand Fair Use try this for a start ( https://www.copyright.gov/fair-use/index.html ) And the reason I bring this up.... anything AI "generates" is in whole or part(s) of existing work(s) - its just what it is.

    I would think at some point here AI generated anything should be tagged as such. There is a more than clear line between "Generated" and "Original" or to throw that into another perspective, there is a clear line between "Fantasy" and "Reality". And when terms like this are used is it not then a reasonable assumption to suggest an end user should be aware of what they are looking at, reading, listening to.

    And then this leads to the question of the OP... should it be removed? I think the more viable answer is it should be identified. And it just so happens that this is the path that has been taken, read here: ( https://about.fb.com/news/2024/02/la...m-and-threads/ ) or here ( https://www.pcmag.com/news/openai-la...easily-removed ) and this article brings up the more than valid point that Image Meta data can simply be removed... then you read ( https://searchengineland.com/google-...-center-437645 ) and I think Google is making it clear that if no Meta Data is present, the image will probably be removed.

    SO now that we understand the landscape of AI "Generated" content as it pertains to its use on Social Media and beyond, we can say that the answer to the question... should it be removed? is already answered for us... NO. it will not be removed, BUT it will be identified as such.
    Signature
    Success is an ACT not an idea
    {{ DiscussionBoard.errors[11791989].message }}
    • Profile picture of the author max5ty
      Originally Posted by savidge4 View Post

      "Generated" is Fair Use spun every which way... BUT Fair Use is not a bad thing... to better understand Fair Use try this for a start ( https://www.copyright.gov/fair-use/index.html ) And the reason I bring this up.... anything AI "generates" is in whole or part(s) of existing work(s) - its just what it is.

      I would think at some point here AI generated anything should be tagged as such. There is a more than clear line between "Generated" and "Original" or to throw that into another perspective, there is a clear line between "Fantasy" and "Reality". And when terms like this are used is it not then a reasonable assumption to suggest an end user should be aware of what they are looking at, reading, listening to.

      And then this leads to the question of the OP... should it be removed? I think the more viable answer is it should be identified. And it just so happens that this is the path that has been taken, read here: ( https://about.fb.com/news/2024/02/la...m-and-threads/ ) or here ( https://www.pcmag.com/news/openai-la...easily-removed ) and this article brings up the more than valid point that Image Meta data can simply be removed... then you read ( https://searchengineland.com/google-...-center-437645 ) and I think Google is making it clear that if no Meta Data is present, the image will probably be removed.

      SO now that we understand the landscape of AI "Generated" content as it pertains to its use on Social Media and beyond, we can say that the answer to the question... should it be removed? is already answered for us... NO. it will not be removed, BUT it will be identified as such.
      I think after looking at the issue from all sides, marking the image as AI generated would probably be the best choice. I'm not an attorney obviously, but this is what I'm thinking. I could maybe have my mind changed by a good argument against marking them.

      I've been hearing stuff on the news about how AI has been wreaking havoc in some instances...

      a principal that was shown on video making racial remarks...which were proven to be AI generated...and the list goes on.

      AI generated images are all over Facebook now. Unfortunately, most can't tell they're AI generated. They're used a great deal for spamming etc.

      Hollyweird is also upset about the whole AI thing.

      I don't think we're at the scary level many of the creators of the big AI programs warned us about yet...

      but I think we're getting close.
      {{ DiscussionBoard.errors[11792001].message }}
      • Profile picture of the author savidge4
        Originally Posted by max5ty View Post

        I think after looking at the issue from all sides, marking the image as AI generated would probably be the best choice. I'm not an attorney obviously, but this is what I'm thinking. I could maybe have my mind changed by a good argument against marking them.

        I've been hearing stuff on the news about how AI has been wreaking havoc in some instances...

        a principal that was shown on video making racial remarks...which were proven to be AI generated...and the list goes on.

        AI generated images are all over Facebook now. Unfortunately, most can't tell they're AI generated. They're used a great deal for spamming etc.

        Hollyweird is also upset about the whole AI thing.

        I don't think we're at the scary level many of the creators of the big AI programs warned us about yet...

        but I think we're getting close.
        If we even begin to think it is scary now.... Right now, AI does what it is told, and does so with the data it is provided. Give AI a script, and some "Voice Lessons" and you can have Mother Teressa cussing like a sailor.... and again with the perspective of "Fantasy" vs "Reality" it is clear that "Generated" content has to be clearly labeled.

        Due to the limitations of perspective in regards to the WF... i will refrain from some more pointed conversation in regards to the direction of AI as it stands.... But for years now... voices have been censored, and that censorship has directly impacted the pool of data. As "Scarry" as AI may or may not be, what is far more terrifying is the powers that be, controlling the data that will influence AI at a later date.

        Ai is only as good as the data that goes in, and the prompting to provide what comes out, and right now it is "Controlled". The Beast - will be unleashed, When Data is truly open, and AI no longer requires prompting... oh nelly!
        Signature
        Success is an ACT not an idea
        {{ DiscussionBoard.errors[11792036].message }}
        • Profile picture of the author Odahh
          Originally Posted by savidge4 View Post

          If we even begin to think it is scary now.... Right now, AI does what it is told, and does so with the data it is provided.

          Ai is only as good as the data that goes in, and the prompting to provide what comes out, and right now it is "Controlled". The Beast - will be unleashed, When Data is truly open, and AI no longer requires prompting... oh nelly!
          When ai starts using the Data it has on an individual and using voice and facial expressions and vital signs to generate an image or song or movie or video game or story.

          With people already withdrawing from each other because anti social ehas gotten much of it's profit giving people plenty of reason and keeping them engaged by battling each other constantly.

          AI gets to come in and restore civility in it's own manipulative fashion
          {{ DiscussionBoard.errors[11792226].message }}
          • Profile picture of the author savidge4
            Originally Posted by Odahh View Post

            When ai starts using the Data it has on an individual and using voice and facial expressions and vital signs to generate an image or song or movie or video game or story.

            With people already withdrawing from each other because anti social ehas gotten much of it's profit giving people plenty of reason and keeping them engaged by battling each other constantly.

            AI gets to come in and restore civility in it's own manipulative fashion
            very utopian... wish this where the case... but Ai can only spit out, what is being put in... it cant "restore" civility if none exist. The way I am seeing this is that AI will simply dumb things down - and as much as that would be far from civility it will without question be quieter? - if that makes sense?
            Signature
            Success is an ACT not an idea
            {{ DiscussionBoard.errors[11792318].message }}
            • Profile picture of the author Odahh
              Originally Posted by savidge4 View Post

              very utopian... wish this where the case... but Ai can only spit out, what is being put in... it cant "restore" civility if none exist. The way I am seeing this is that AI will simply dumb things down - and as much as that would be far from civility it will without question be quieter? - if that makes sense?
              Yes quieter. Rather than two very different people constantly arguing because they have very different tastes. Bogging down any discussion. People are different.

              As ai gets better over the next few years and more responsive to individuals.so that ai generated art piece might be generated as much by data the ai has on the person writing the prompts.

              If that person posts that on social media for people to see. The ai and algorithms will probably limit anyone with harsh criticism. Or who is just being mean from being able to comment.

              The mods do it here and I see several of my YouTube comments disappear within seconds of me hitting enter and I'm not even being critical.

              Going back to the op a second. The occasional posting of ai generated content by an individual is fine. Large scale spam though not so much.
              {{ DiscussionBoard.errors[11792337].message }}
              • Profile picture of the author savidge4
                Originally Posted by Odahh View Post

                Yes quieter. Rather than two very different people constantly arguing because they have very different tastes. Bogging down any discussion. People are different.

                As ai gets better over the next few years and more responsive to individuals.so that ai generated art piece might be generated as much by data the ai has on the person writing the prompts.

                If that person posts that on social media for people to see. The ai and algorithms will probably limit anyone with harsh criticism. Or who is just being mean from being able to comment.

                The mods do it here and I see several of my YouTube comments disappear within seconds of me hitting enter and I'm not even being critical.

                Going back to the op a second. The occasional posting of ai generated content by an individual is fine. Large scale spam though not so much.
                Doing my best here to refrain from a political blow out.... but what you are describing AI will be doing is filtering and Censoring.

                So within "Free Speech" there are 5 known limitations ( incitement, defamation, fraud, obscenity, child pornography, fighting words and threats ) this goes back to a Supreme Court finding in Brandenburg v. Ohio - 1969 - so no I am not making this stuff up.

                I kinda keep saying this but AI is in the process of dumbing things down... and I think to a point thats fine... the issue I have is the Bias in which the dumbing down is occurring.... equally and straight across the board within the 5 known legal limitations - yep all for it - have at it - the world will be a better place. but get out of that... and you can call it what you will, and I will just call it wrong

                Now in regards to your YouTube posts being removed... as an Author of material you can create a list of catch words when used in the comments, the comments are removed "instantly" same is true with Facebook group commenting. So its not Facebook or YouTube specifically blocking you, it is the creator of the content that is in essence blocking you.

                The other piece in this is "Section 230" that removes the liability of these platforms for the views and opinions expressed on the platform by 3rd parties ( end users ) - basically you cant have it both ways And again, apply filters and censorship to the 5 known limits and I simply have no issue. The WF being privately owned and I would suggest and having a very well thought out and written set of rules of what can and can not be said.. and I cant say i have ever heard of an instance where anyone on the WF mod staff has blatantly just censored someone just because...

                You want to use AI thats great - i can in no way say that i dont use it - but it is becoming clearer by the day that there needs to be a clear and defined usage and allowance as it applies to the laws of the land. How many times do you have to hear people with Authority complaining X ( Musk ) would not comply with a request to Censor... before you understand that if you are only hearing about X... the others are complying?

                case and point here... what 5 yrs ago some mad man ( oh the gaul of this guy ) was saying hey lets shut down TikTok for national security reasons? a vast majority of the conversation back then agreeing with the idea... was CENCORED - removed, gone, bye bye... and here we are July 28th TikTok will be no more... HMMM

                Recent history is not fairing well for the ideas and concepts of filtering and censoring there are simply a whole lot of "Conspiracy Theories" that over an amount of time have proven to be correct. There simply is a track record of maybe censoring the bad stuff... but there are many times the bad stuff that is being censored, is really the truth - and that is simply dangerous.
                Signature
                Success is an ACT not an idea
                {{ DiscussionBoard.errors[11792351].message }}
  • Profile picture of the author Monetize
    Artificial Intelligence (AI)

    Plus (+)

    Chicken Little

    Equals (=)

    The Sky is Falling

    OMG

    AI is going to take over the WORLD!!!

    FYI

    Using AI for content creation is no different than using a

    GHOSTWRITER.

    Except I don't have to wait two weeks for delivery.

    The images it produces are original enough.

    Did AI invent yellow, red, or blue?

    Obviously it's going to use the same

    COLORS

    That Leonardo Da Vinci, Renoir, Michelangelo, Monet,
    Van Gogh and everybody else ever used.

    The Bible (Ecclesiastes 1:9) says

    There is nothing new under the Sun

    But of course there's something new, specifically AI that
    the OP is referring to is relatively new, it certainly was not
    in existence when The Bible was written.

    If I ask ChatGPT a complex question like:

    ChatGPT, I am going to combine STEP 5 with STEP 12,
    flip it sideways and stand it on its head, what will the
    outcome be?

    ChatGPT's output will be original content because that
    question has never been asked of it before.
    {{ DiscussionBoard.errors[11792037].message }}
    • Profile picture of the author max5ty
      Originally Posted by Monetize View Post

      Artificial Intelligence (AI)

      Plus (+)

      Chicken Little

      Equals (=)

      The Sky is Falling

      OMG

      AI is going to take over the WORLD!!!

      FYI

      Using AI for content creation is no different than using a

      GHOSTWRITER.
      I can only speak for myself and the comment I made.

      I wasn't referring to AI being scary for those working out of their house and doing some type of marketing to try and make a buck or two.

      I'm speaking of AI as a whole.

      Apparently, I'm not the only one that says it has the potential to be scary...

      I would point to exhibit A...the U.S. Government has also raised concerns and has formed a committee to look into it.

      Exhibit B...many countries around the world have voiced concern and have begun conversations about its capabilities.

      Exhibit C...those who created the AI platforms have raised concerns about its potential...even putting out a letter that was a warning signal.

      So, using AI to do your blog posts or whatever wasn't the thought behind my comment about its drawbacks.

      I, as many around the world, are looking at it based on its potential.
      {{ DiscussionBoard.errors[11792038].message }}
  • Profile picture of the author savidge4
    Now that we have attempted to dumb this down...

    Color... sure Leonardo Da Vinci, Renoir, Michelangelo, Monet,
    Van Gogh used color... prehistoric man by default drawing on cave walls used color... again sure ok fine.

    BUT could any of those examples replicate a color? and when I say replicate I mean exact? in todays world... depending on the color spectrum you use there are millions of colors... RGB as an example there are 16,777,216 colors to be exact. and at any given moment you can replicate a color used - EXACTLY. we have unparalleled accuracy, as compared to our predecessors. it wasnt that long ago we had access to 256 colors... and now, we can replicate the full spectrum of color. Digital is different.

    The day will come.... that the current language of choice be it Binary (00000000 to 11111111) or hexadecimal (00 to FF) will be a thing of the past - There are simply to many limits... case and point there are only 16,777,216 colors. THIS will increase exponentially.

    We are in the early era of the Digital revolution... and to understand what that will look like i the future, we can look to the past... we can look at the mechanical revolution. The mechanical revolution replaced muscles with machines. There are people on this earth today that were alive when "Horseless Carriage" was a thing - the Muscle power of horses were replaced with the mechanics of cars. Look at were we were in say 1900 vs now mechanically? its mind boggling.

    The digital revolution has been taking place for 50 years now? we have more power in our hands ( phones ) than a room of electronics had back in the day.

    So what in part ( or whole ) is driving this revolution? The answer is DATA... easy enough, so then the question to ask is where is this data? Its easy to answer this by saying "the internet" but that is a real shallow answer... I have said for years now... maybe even years and years... Google is not a search engine company, they are in the business of collecting data - and I might add at a scale I dont think we can quite comprehend. Google Maps is mind boggling... Every sing word from every single book in the Library of Congress has been scanned by Google... the data collected from any and every Android phone - and the extent of that data capture I am sure is far greater than we can even begin to assume. Google itself probably has all the data needed to solve any and every human problem present and unknown.

    Someone that is probably the most public figure for AI... Elon Musk sees the writing on the wall... understands the path AI is taking. Ponied up 44 BILLION dollars to stay in the game and maintain an amount of influence. Musk did NOT buy Twitter to save free speech... he bought Twitter because of all the AI elements the one part and probably most important part he was missing was LANGUAGE, and more importantly language that is devoid of filtering, censorship, and bias.

    I would ask who has played with Grok... but i know there might be 3 of us that could answer yes to that.... GROK, is a LANGUAGE data base, and more importantly an OPEN SOURCE language data base. Those with programming knowledge currently have access to the cleanest least filtered least censored least biased platform of data.

    To maybe better understand this think about the Internet in China... I dont think any of us will disagree that it is highly regulated, censored and Biased. what would ChatGPT look like running that set of data? it would be silly. and then you have ChatGPT as it is here in the States... and then think what it might be like running off of the GROK data. I can personally tell you - Its night and day. I keep saying this, the Google data sets we have access to are flawed, and I would say less by default and more by intention. There is not a doubt in my mind Google can filter the flaws. AI is only as good as the data it has access to.

    The AI we currently have access to is pretty much dumb as a box of rocks. At this point... much like computers in general, it is only as smart as the person in front of it. Look at me I have proprietary prompts is laughable... start filtering data through 2 3 4 8 different language filters before you "prompt" and then you are getting somewhere.

    And the whole there is no difference between GPT and a Ghostwriter? oh my... I would never ask GPT to write an article for say a law firm ( as an example ) but I would ask a law professor.

    And at the more Macro level... AI written material is... well has become... the bare minimum at this point. Go look at the SEO section... "my stuff is not indexed" and you get into it and the articles are poorly written and duplicate.... those days are done... The MINIMUM is some precontrived spew by AI, and anything less than that is no longer indexed... in many ways its a good thing, and I in no way knock the whole idea... but I personally dont shoot for minimums - ANYBODY can reach the minimum... its the 5% that go beyond that and introduce knowledge and experience into the writing.
    Signature
    Success is an ACT not an idea
    {{ DiscussionBoard.errors[11792064].message }}
  • Profile picture of the author spartan14
    Well i get your point but why social media dont need to use ai images but websites use AI content ?
    Getting to your point it should that nothing should use ai content but its used more and more
    So i think we should accept it as part of our marketing strategy etc
    {{ DiscussionBoard.errors[11792090].message }}
  • Profile picture of the author CyberSEO
    I will tell you more. Stack Overflow won't allow you to post anything generated by ChatGPT because... they give their data to OpenAI to train their models, so they don't want to train them on the content generated by AI. What an irony )
    Signature
    AI Autoblogger - the ultimate autoblogging WordPress plugin, powered by OpenAI GPT-4, Anthropic Claude, Google Gemini Pro, Mistral Large, Midjourney, DALL-E 3 and Stable Diffusion XL
    {{ DiscussionBoard.errors[11792126].message }}
  • Profile picture of the author a8
    where i can find for learning about AI create Image


    [Sig file violation. English language only - please refer to forum rules]
    {{ DiscussionBoard.errors[11792650].message }}
  • Profile picture of the author Kausik Biswas
    One can manually generate image. Also can take help of AI. But the created image must be crosschecked, whether it is unique and actually helpful. Creator must use prompt engineering properly. If so, then I think it can be used in social media too.
    {{ DiscussionBoard.errors[11793763].message }}
  • Profile picture of the author IM Simple
    In my opinion, AI images have their own advantages and disadvantages. It makes the work easy for businesses or entrepreneurs who want to improve their businesses. However, it's like disrespectful to real artists out there as well as to their natural talents.
    {{ DiscussionBoard.errors[11793955].message }}
  • Profile picture of the author NillBraun
    The problem with AI-generated images is that they do not reflect the personal experience or creativity of the user, and as a result, they lose their authenticity. Social networks should maintain a balance between technological innovations and the authenticity of user-generated content.
    {{ DiscussionBoard.errors[11793997].message }}
  • Aw, see here how Taylah Swift dressed up as Scarlit Yannosum to kinda set whatevah in motion?

    Or was it Starlit Supposesum who lit up Maybe're Drift as a meme you might wanna tap inta?

    Do my distinctions buttween The Taylstah an' Whack! Ditto! surprize or 'maze ya?

    Gotta figure this is less to do with MOI an' yr own cuppassity to distinguish Princess-indooced ROT frahm ACTSCHWL ROT than mebbe immediate to yr eye-2-brain co-ords rn.

    See, bcs, in our hearts an' minds an' lives, we thrive so vurrmuch bettah on our exchanges of enthoosiasm an' authenticity that we do 'pon artifice destined to invite only harm.

    Gaht no ideah musself what it takes to make Ms YoHaveswumm's tits look joocieah than mine, in that technopersuazo kinda way AI's hellhole compels one an' all now to PLUMMET SUPAHDOWNO, but less'n we can compare mortal zistoes we gotta measure out on virtyool improbables.

    Or did I say Swollen Thrift an' Smylat Allglancewon?

    tbh, without tangible refrnce, all you gaht are boats driftin' out noplace on a doomwish.
    Signature

    Lightin' fuses is for blowin' stuff togethah.

    {{ DiscussionBoard.errors[11794001].message }}
  • Profile picture of the author John Daniel M
    Where I Can Find Create Animate AI Images? Is there Any App or Site To Generate
    {{ DiscussionBoard.errors[11794038].message }}
  • Profile picture of the author Rosewatt
    I agree that AI generated image are not authentic. If not carefully used then it might cause a serious trouble in future.
    {{ DiscussionBoard.errors[11794400].message }}
  • Profile picture of the author ananittarhib
    It's understandable that you're worried about the reliability of user-generated material when pictures made by AI are also present. User-generated content (UGC) is very important because it provides real conversations and personal experiences on social media. Artificially intelligent pictures can help people be more creative and give them new tools, but they shouldn't take away from the human touch that makes social media interesting and approachable.

    Sincerity is very important, and if people think the content isn't real, it could hurt trust and interest. Smart computers can help and improve things, but real people talking about their own experiences should still make up most of the material. By doing this, we can get the best of both worlds: the creativity and help of AI and the honesty and relationship of UGC.
    {{ DiscussionBoard.errors[11794519].message }}
Avatar of Unregistered

Trending Topics