Despite the apparent demise of blogs the flat line of the RSS-ograph blips with a pulse from David Kernohan “on chatbots.” FOTA is alive!
Unsure if my comment gets through the gate (a first one generated a critical WordPress error, sorry, David), but I have to at least assert my assertion, as if it blips anywhere in the raging discordant discourse, “Intelligence might be based on pattern recognition as Stephen [Downes] asserts, but it should not be mistaken for intelligence.”
So when David passes a linked reference to the Colossus as the dawn of pattern guessing to decrypt war time messages, my pattern recognition goes to where no GPT can fabricate:
This photo was taken on my own visit to the National Museum of Computing in Bletchley Park, that being a memorable day when Dave and his partner Viv drove me all the way from Bristol where I visited them to Milton Keynes where I spent a week at the Open University.
Maybe a machine could mine the facts from my blog posts and photos, but it would never make connections, the feelings, to the experience of being there that are not digitized or accessible to wholesale scraping. Never.
Or is this my own flailing effort to raise a pitifully tiny flag of I am Human in front of the advancing, inevitable horde of machines? For an image I could have plopped a prompt into a DALL-EMidJourneyStable Diffusion but why, when I can deploy one of my own making?
I could try my best to weave more words around my emerging thought patterns, yes ones that I generate from my own sum of vast experiences. And truly, I could say that I myself, with this nerve network plugged into a 3 pound skull enclosed non-battery powered device, merely have been training 50+ years on written, visual, auditory media,much of which I did not ask explicitly to use, from which I generate through some mystical process, my “own” words? my “own” imagery?
Everything is a Remix but AI Does Not Remix Like an Artist
Who better to turn to than Kirby Ferguson to wisely delve into Artificial Creativity?
Stop, watch the whole thing. I mean the whole damn series. I can only yank quotes
Of all Humanity’s technological advances, artificial intelligence is the most morally ambiguous from inception. it has the potential to create either a Utopia or a dystopia. Which reality will we get? Just like everybody else I do not know what’s coming but it seems likely that in coming decades these visions of our imminent demise will seem campy and naive because our imaginings of the future always become campy and naive.
Everything is a Remix Part 4
He takes AI to “court” on three counts, and makes a point that many don’t want to accept, that harvesting all of the “stuff” readily available is maybe not the point of ethics to hang the purveyors. If you buy into his theme that everything is a remix, that means everything is available, as he has done in his video.
But do not take this as suggesting there is a free ticket to just grab content for the classic “because you can” reason. Follow Kirby Ferguson’s statement about all the media he has remixed into his video:
On some videos about AI the big reveal is that this video was actually made by AI. But this video and this series is the opposite. Nothing has been AI except where I cited AI art. This is entirely human made, the words are all mine but they’re merged from the thoughts of countless people. Everything you’ve seen and heard is from real filmmakers and musicians and game developers and other artists. All these thoughts and all this media were remixed by me into something new and yes I did it all without permission.
Everything is a Remix Part 4
The big difference is that this filmmaker provides credits / attribution to he sources. It is very clear what was used. There is no mask of source content or how it was used hidden behind a facade of a commercial purveyor whose very name has washed open with techno-clorox.
Also, lost in the court section is a very valid question-
Training AIs on individual artists work does seem wrong everyone should be able to opt out of all training sets and maybe AIS should simply not train on images from active art communities. Also some company should make an image generator trained on public domain and licensed images which would avoid this Hornet’s Nest entirely. Somebody please do this.
Everything is a Remix Part 4
Why is there no ethical entity out there creating training from public domain or openly licensed materials? Or why does quote/unquote “OPEN” ai DOT com, which already trains its machines on Wikipedia amongst everything else, just create a version limited to truly open content? About the only thing I found was an image generator on hugging face that looks like it does this, but I am not clever enough to make it do anything.
There is a free idea for anyone to pick up.
Finally, Kirby Ferguson ends with a compelling (to me) assertion of the essence of creativity.
AIs will not be dominating creativity because AIs do not innovate. They synthesize what we already know. AI is derivative by design and inventive by chance. Computers can now create but they are not creative. To be creative you need to have some awareness, some understanding of what you’ve done. AIs know nothing whatsoever about the images and words they generate.
Most crucially, AIs have no comprehension of the essence of art, living, AIs don’t know what it’s like to be a child, to grow up, to fall in love, to fall in lust, to be angry, to fight, to forgive, to be a parent, to age, to lose your parents, to get sick, to face death. This is what human expression is about. Art and creativity are bound to living, to feeling. Art is the voice of a person and whenever AI art is anything more than aesthetically pleasing it’s not because of what the AI did it’s because of what a person did.
Art is by humans for humans.
:
Everything is a Remix is a testament to the brilliance and beauty of human creativity. In particular it’s a testament to collective creativity. Human genius is not individual it is shared.
Everything is a Remix Part 4 (emphasis added by me)
Please watch this video! All of them!
Back To The Hammer Hand
But it’s not as clean as just going John Henry and making an untenable slice of human versus machine. Artificial Intelligence “stuff” is a tool, but it’s not “just a tool.” I am reaching back to something I often rely on from Gardner Campbell’s explanation of Marshall McLuhan
“There is no such thing as “just a tool.” McLuhan wisely notes that tools are not inert things to be used by human beings, but extensions of human capabilities that redefine both the tool and the user. A “tooler” results, or perhaps a “tuser” (pronounced “TOO-zer”). I believe those two words are neologisms but I’ll leave the googling as an exercise for the tuser.
The way I used to explain this is my new media classes was to ask students to imagine a hammer lying on the ground and a person standing above the hammer. The person picks up the hammer. What results? The usual answers are something like “a person with a hammer in his or her hand.” I don’t hold much with the elicit-a-wrong-answer-then-spring-the-right-one-on-them school of “Socratic” instruction, but in this case it was irresistible and I tried to make a game of it so folks would feel excited, not tricked. “No!” I would cry. “The result is a HammerHand!”….
http://www.gardnercampbell.net/blog1/doug-engelbart-transcontextualist/
So no “just a tool,” since a HammerHand is something quite different from a hammer or a hand, or a hammer in a hand. Gardner has given me more directly, in email:
I got to that in part because of McLuhan’s famous dictum “the medium is the message.” Most folks appear to think he meant that the medium shapes the message. If you read the piece in which the phrase appears, however, you can see that’s not what he meant. Instead, McLuhan thought of every medium as a message about what we are and desire as human beings. He said the electric light was a message. Every medium should tell us something meta about itself, and something vital about humanity. A medium is not just a channel for transmitting stuff. A medium is also itself a message, a transmission. Can we understand the medium’s message about itself, and thus about us? That’s why the book is called Understanding Media. What is the message these media convey about themselves? and about mediated experience generally?
So with that, I built on Alan Kay (and I think others as well), who said “we shape our tools, and after that our tools shape us,” bringing in the idea of man-computer symbiosis, putting it all within the context of Engelbart’s integrated domain, and then re-reading McLuhan to find a way to express what I took to be something essential about his ideas of human transformation in the development of mediated experience, and I came out with hammerhand.
Gardner Campbell, personal communication
Much of the educator reaction to ChatGPT (which to me is narrow as there is much more we should be wrapping our heads around), so focused on the fear/worry/change factors rather than ” ideas of human transformation in the development of mediated experience.”
So This Thing Happened Today Which I Defy Anyone To Experience By Typing Prompts Into a Box
Going back to where I started, with David Kernohan’s On Chatbots post, he gives just a short bit at the end to maybe the larger idea of his two, under the heading “A Matter of Semantics”:
I want to close my argument by thinking about the other major strand of artificial intelligence – an associative model that starts (in the modern era) with Vannevar Bush and ends with, well, Google search. The idea of a self-generating set of semantic links – enabling a machine to understand how concepts interrelate – is probably closer to the popular idea of artificial intelligence than toys like ChatGPT.
http://followersoftheapocalyp.se/on-chatbots/
meaning (I think) that the interconnected web of ideas imagined by Bush that influenced Engelbart and actually was instantiated by Tim Berners Lee, is the connectivist idea that the web itself, changing with every new bit linked on to it, offers more potential for making something akin to intelligent than chatbots that are merely regurgitation parts of it in a way that just parody intelligence, not embody it.
So this happened today. It is of no significant to any discussion threading out in the ghosted public square of twitter or the de-aggregated butvibrantcorners of Mastodon, certainly not dead to ne where I will never vention spew pots of (f*** it I cannot call it “Meta” its always Facebook),or the iteration of the America Online is to the real Internet as Linkedin is to ______________… Oh I might have lost my thought completely, as it humanly happens.
Can I get help?
This is what people have their undergarments wadded about? Seriously? I push back
Okay, I am left to my own story making.
Today I sat down to catch up on a few DS106 Daily Creates, it being the very essence of acts of human creativity assisted by tools (using a “TDC HAND”). This was one challenge from a few days ago which in true TDCness, gives a nudge, and opens a door to respond in almost any way.
Lots of ways to run with this, so I just start with the random names generator that suggests possible names from different countries. Cool! I love random stuff and never saw this one. There’s 12 countries listed, each with 10 names. I just decide to be quick and use the first names in the middle row:
Before getting to where / how they meet, I decided I need pictures. Before everyone got wrapped up in Generative text posing as intelligence, there was the phase of all the generative adversarial network (GAN) tools making realistic photos of people and thing that do not exist. If you want a real diversion, see This X Does Not Exist. But I went for the one I remember for generating people, thispersondoesnotexist.com but that now seems gone and only goes to some AI outfit.
But I did find a similarly-URL-ed version at https://this-person-does-not-exist.com/ that was interesting,as there are a few more options to choose from (gender,age range, a few ethnicity options, so I generated 4 non-existent people for Ionut, Lázár, Angel, and Elenor. I imported into Photoshop using one of the Panorama collages which spread them out like photos on a table.
Then I tried to think if where to place these non-existent people. I first reached for a new browser window thinking of some sort of technical image, like a computer circuit board. This is when unexpected-ness happened.
You see I use the Library of Congress Free to Use Browser extension that puts a random public domain image in my screen each time I open a new browser tab. I was fully intending to open an image search, but there, but random chance here was my answer, a road sign for Hanks Coffee Shop, even better, because it was from Bensen Arizona, a place I have been before.
So now it all came together, these people who do not exist, met up for coffee at Hanks in Benson. A bit more Photoshop editing to make a cloud background, superimpose the names of the four, and I was done.
“So what?” is certainly a reasonable response. Couldn’t I save time and just type into an image prompt box, “Photos of 4 people displayed under an old time coffee shop sign”? And maybe iterate a few times until it’s “good enough”? Yes, but is making art about the process or the product? Maybe sometimes it is just getting the thing done, turn it in, as they say.
But what is the connection to it? Would an AI remember driving through Benson, AZ on a memorable road trip to camp in the Chiricahua mountains? Would it remember a completely un-related connection from these photos in the Flickr Commons and that there was a call a while ago for examples of galleries of themed images from the commons? And would it then decide, for no productive reason, to search for other Arizona road sign images in the flickr commons, create a gallery, and then share it back?
I’d say, plausibly, eff no. I want to be doing stuff described as “Art is the voice of a person and whenever AI art is anything more than aesthetically pleasing it’s not because of what the AI did it’s because of what a person did.”
I’m not saying at all don’t do AI. And I absolutely intrigued by what it might offer, we have hardly even scratched the surface. But it does not always mean we have to just line up as robot servants to the AI Industrial Complex.
If we lose our ability, interest, to be non stochastically human in our tasks, then we end up being “derivative by design and inventive by chance.”
Fin
Never. But I am hoping maybe to see before not too long, another just thinking blip from FOTA.
Featured image: Humanly created and selected, mine
I was happy to see Kernohan thinking aloud on FOTA again too 🙂