I disagree that generative AI is a novelty item in general; art AIs are already being used in prototyping, background creation, etc. and we're seeing entire products created using AI generated art. Lower print run products cannot afford to pay artists tens of thousands of dollars to hundreds of thousands of dollars to produce a fully ill…
I disagree that generative AI is a novelty item in general; art AIs are already being used in prototyping, background creation, etc. and we're seeing entire products created using AI generated art. Lower print run products cannot afford to pay artists tens of thousands of dollars to hundreds of thousands of dollars to produce a fully illustrated high quality full-color RPG manual, but with MidJourney, you can make 99% of the art that way, pay someone $500 to draw you cover art, and you've got what you need for $530 instead of $50,000. This is a big improvement. Creating art for TTRPGs is another big thing that it's really useful for. Stock images can be replaced with a MidJourney subscription.
Moreover, because art doesn't have to be "true", it just has to look good, as well as because something that looks cool can still be used (the Clever Hans effect), all the AI art has to do is produce generally the right sort of thing, and make stuff that looks really good, to be a "success".
We're also seeing it being used in photoediting software; photoshop combined with generative AI art is a very powerful combination and is very useful for editing photos, removing foreground objects, adding stuff in, etc.
We're also seeing AI upscaling being put to good use, as well as outpainting.
I think the language models are apt to be a bit less useful outside of some of the more edge-case uses as far as productivity goes, as chatting with them is a novelty and they can't write good enough material (and can't consistently enough avoid hallucinations) to be nearly as useful as they need to be for a lot of purposes.
That said, they may be useful for translation tools, as well as generating material for video games - I've already seen people use these models to generate NPC dialogue and a demonstration of using it (along with a voice synthesizer) to create sports casting that is responsive to what is going on in a video game race.
Hard disagree. Art and design are communication, like language. Saying that the purpose of visual communication is to make things that "look good" is like saying that writing is about making up stuff that "sounds true".
It's an evolving form of human expression, like language. Behind every famous painting is a story to be told of its time of creation, subject, themes, the story of its author.
On a different note, pictograms are one of the most common forms of communication in society. They are so ubiquitous, we barely notice them. From traffic signs to phone icons.
The fact that an AI can replace the creation of MEANINGFUL art (I use the word without sentimental value attached, "meaning" is used here to mean something that was created with a goal, to communicate something) with something that simply looks aesthetically pleasing without meaningfully contributing to the human experience. An AI could not have come up with the familiar little figures you see on every WC.
Yet using AI to replace meaningful art and design work disincentivizes meaningful communication and advancement in that field. The effects of this are much more abstract and difficult to measure that the mass proliferation of hallucination-fuelled fake news, but we can all tell a soulless cash grab of a Marvel movie from an artfully created one that inspires something deep within us.
I fear that even the realm of art and entertainment, which is a refuge from the cold, numbers-driven world, will become another desolate, meaningless void without a human element, shunting out the artists who used to contribute to the evolution of visual communication by denying them a fair living.
I disagree that generative AI is a novelty item in general; art AIs are already being used in prototyping, background creation, etc. and we're seeing entire products created using AI generated art. Lower print run products cannot afford to pay artists tens of thousands of dollars to hundreds of thousands of dollars to produce a fully illustrated high quality full-color RPG manual, but with MidJourney, you can make 99% of the art that way, pay someone $500 to draw you cover art, and you've got what you need for $530 instead of $50,000. This is a big improvement. Creating art for TTRPGs is another big thing that it's really useful for. Stock images can be replaced with a MidJourney subscription.
Moreover, because art doesn't have to be "true", it just has to look good, as well as because something that looks cool can still be used (the Clever Hans effect), all the AI art has to do is produce generally the right sort of thing, and make stuff that looks really good, to be a "success".
We're also seeing it being used in photoediting software; photoshop combined with generative AI art is a very powerful combination and is very useful for editing photos, removing foreground objects, adding stuff in, etc.
We're also seeing AI upscaling being put to good use, as well as outpainting.
I think the language models are apt to be a bit less useful outside of some of the more edge-case uses as far as productivity goes, as chatting with them is a novelty and they can't write good enough material (and can't consistently enough avoid hallucinations) to be nearly as useful as they need to be for a lot of purposes.
That said, they may be useful for translation tools, as well as generating material for video games - I've already seen people use these models to generate NPC dialogue and a demonstration of using it (along with a voice synthesizer) to create sports casting that is responsive to what is going on in a video game race.
Hard disagree. Art and design are communication, like language. Saying that the purpose of visual communication is to make things that "look good" is like saying that writing is about making up stuff that "sounds true".
It's an evolving form of human expression, like language. Behind every famous painting is a story to be told of its time of creation, subject, themes, the story of its author.
On a different note, pictograms are one of the most common forms of communication in society. They are so ubiquitous, we barely notice them. From traffic signs to phone icons.
The fact that an AI can replace the creation of MEANINGFUL art (I use the word without sentimental value attached, "meaning" is used here to mean something that was created with a goal, to communicate something) with something that simply looks aesthetically pleasing without meaningfully contributing to the human experience. An AI could not have come up with the familiar little figures you see on every WC.
Yet using AI to replace meaningful art and design work disincentivizes meaningful communication and advancement in that field. The effects of this are much more abstract and difficult to measure that the mass proliferation of hallucination-fuelled fake news, but we can all tell a soulless cash grab of a Marvel movie from an artfully created one that inspires something deep within us.
I fear that even the realm of art and entertainment, which is a refuge from the cold, numbers-driven world, will become another desolate, meaningless void without a human element, shunting out the artists who used to contribute to the evolution of visual communication by denying them a fair living.