Just using the term AI is not helpful in the discussion.If a creator uses AI they should expect to be replaced by AI
Also, why would someone using AI to generate content expect to be paid for the content?
This defeatist attitude is really what says it all.The people who were most bullish on AI were, if anything, the least optimistic about their own career prospects. “I think at a certain point it won’t matter,” Kavan Cardoza told me. “It’ll be that anyone on the planet can just type in some sentences” to generate full, high-quality videos.
This is the talk of a person who doesn't value his own contributions to the medium, and perhaps should be doing something where he has more to offer.This defeatist attitude is really what says it all.
They are new, short dramas shot and edited for vertical video. The business model tends to be free episodes followed by $1 or $2 premium episodes. This episodewhat the hell is a "vertical drama"?
You're not going to sway me with "I can't afford to do it otherwise." Sorry buddy, then don't do it. Labor cost is a thing everyone has to deal with. Don't you think amazon and mcdonalds and walmart would love to replace their entire workforce with robots and/or slaves they don't have to pay? Just because you don't/can't want to pay someone doesn't give you the moral high ground to claim AI is good.
It turns out Quibi was just too early and not the terrible idea it was on paper.They are new, short dramas shot and edited for vertical video. The business model tends to be free episodes followed by $1 or $2 premium episodes. This episode
of The Ankler is a great primer on vertical dramas.
I'm so annoyed by this, AI generated video didn't exist in 2016. Miyazaki was shown a video of an ML algorithm crawling around unnaturally with a monster 3d model and was, for some reason, reminded of a disabled friend.In 2016, the legendary Japanese filmmaker Hayao Miyazaki was shown a bizarre AI-generated video of a misshapen human body crawling across a floor.
How does this work? If I am a stock image creator I assume I will get paid every time someone pays to use it. But AI is only trained once (so it only uses the image once) and then it can forever generate similar content. Is there a higher fee for creators who agree to have their images used for AI training?“feels pretty good” using Adobe products because the company trains its AI models on stock images that it pays royalties for.
Less than useless, an active detriment to the usefulness of search engines. Even specifying -AI doesn't work unless it's all correctly tagged (spoilers: it's not).Ai generated photos and video are a scourge and should be considered artless. Using these tools to replace workers wholesale is something we should outright consider unacceptable. A truly abhorrent example of media complicity.
AutoCAD makes things that become real. Its entire purpose is to help people organize things in a rational, efficient way so that they can be made into real objects reliably. Generative AI is all hallucinated fantasy from top to bottom. The only reason it exists is to pump up companies' valuations by selling the fantasy that it can be a useful tool beyond making low-quality fakery.As an architect, this reminds me a lot of the transitions from hand drafted drawings to autocad and now BIM. It is just similar in tone. Despite the rhetoric on both sides, what has ended up happening is that we all use CAD tools, but our workflow and output is still organized as if we are drafting.
I suspect something similar will happen with AI generation. We will see work still created by live action filming, but many scenes will be supplemented or replaced by ai generated content. So the era of a giant set may be over.
I'd rather Mcdonalds and Amazon use robots than exploit humans by paying them less than they need to live.You're not going to sway me with "I can't afford to do it otherwise." Sorry buddy, then don't do it. Labor cost is a thing everyone has to deal with. Don't you think amazon and mcdonalds and walmart would love to replace their entire workforce with robots and/or slaves they don't have to pay? Just because you don't/can't want to pay someone doesn't give you the moral high ground to claim AI is good.
Cheating jerkass rubber-banding ass jackals, one bad corner after a whole race and they're right there!Do people also hate Euphoria ragdolls in Rockstar games or most bots in racing games?
I will say "AI" denoise models are incredibly impressive (I've used adobe and topaz) and have been a great help for some of my noisier photos.Just using the term AI is not helpful in the discussion.
Using a plugin to remove noise (that is sold as "AI/ML" derived) is just doing the job like other plugins did before it, better or worse, and they always had algorithms to do their tasks.
Replacing real, shot footage with AI-generated one, on the other hand (and then not even disclosing that), is another story... and get the f out of here with AI-gen (personalized/targeted) advertising.
Slight tangent, but turns out the real trick with FEAR is that the enemies are just programmed to shout out whatever you are doing. They don't necessarily know how to effectively counter you hiding behind a couch or how to counter flank you. But if you have in game soldiers shout out "he's behind the couch" or "he's flanking us" it tricks the player into thinking there's more going on there than there is. Plus the level design of these games being mostly empty office buildings and warehouses just lends itself well to bot pathfinding.Cheating jerkass rubber-banding ass jackals, one bad corner after a whole race and they're right there!
Seriously though, it's a good point and unfair to put ML in the same basket as generative AI. I'm playing FEAR again and it still has some of the best combat AI, and obviously nothing to do with ChatGPT and its ilk.
Seems relevant to this discussion, IMO.Which is to say: FEAR's best trick is making it feel like it has really smart AI, which turn out to be more important than actually having super smart AI.
The transition from hand drafted to CAD design is more equivalent to the rise of CGI special effects and scenes in film than what is happening with ai generated content. CAD and CGI both digitized the workflows of skilled, talented professionals. AI generation seeks to eliminate those workflows entirely by copying the published works of all those professionals without compensation. I'm actually quite surprised you are not more concerned. As soon as a building construction company can convince a government building inspection body that AI generated plans are safe professional architects will start losing their jobs.As an architect, this reminds me a lot of the transitions from hand drafted drawings to autocad and now BIM. It is just similar in tone. Despite the rhetoric on both sides, what has ended up happening is that we all use CAD tools, but our workflow and output is still organized as if we are drafting.
I suspect something similar will happen with AI generation. We will see work still created by live action filming, but many scenes will be supplemented or replaced by ai generated content. So the era of a giant set may be over.
You may be right but based on the current state of the art that won't be happening anytime soon (I hope!)As soon as a building construction company can convince a government building inspection body that AI generated plans are safe professional architects will start losing their jobs.
Coffee shop artist aka barista![]()
I'd rather Mcdonalds and Amazon use robots than exploit humans by paying them less than they need to live.
"Production Artist" is the term that I'd use... I was one at the start of my career, nothing to be ashamed of.“is a YouTube thumbnail artist.”
Thumbnail creator or thumbnail technician perhaps. But artist?
Excretors.I can't feel good calling about the people using AI for this sort of thing "creators". I would suggest the alternative label "sloppers". Seems more accurate about who they are and what they are actually doing.
Agreed. There will be niche audiences for something more authentic, but most will be fed 'good enough' AI. After all, most of the world seems to find the ad soaked internet preferable to subcribing to what is valuable. One can hope for a revival of community theater perhaps? Or maybe some Nick Park eccentrics who lean in to difficult time consuming ways to produce unique visual experiences. The mass market is hosed for exactly the reason you point out.Effortless creation is inherently caustic to anything that requires our time to be enjoyed. Economic incentives will turn everything AI touches into a race to the bottom, a flood of endless quantity-over-quality slop desperate to extract any return on investment by parasitically devouring our attention.
In the AI future, everything is ads and we'll all treat it accordingly.
That's of course ignoring the fact that it isn't actually good enough to do much of the above, and between the general stagnation and staggering losses it operates at that doesn't look likely to change any time soon.
It works much the same as a sweatshop does. Yes, you can legally (or quasi-legally) employ people who live in an area with such poor prospects that the pittance you give them is one of their best opportunities not to starve. Until we ever reach post-scarcity, there will always be such places.How does this work? If I am a stock image creator I assume I will get paid every time someone pays to use it. But AI is only trained once (so it only uses the image once) and then it can forever generate similar content. Is there a higher fee for creators who agree to have their images used for AI training?