←back to thread

Adobe deletes Bluesky posts after backlash

(petapixel.com)
553 points bookofjoe | 2 comments | 11 Apr 25 14:01 UTC | HN request time: 0s | source
Show context
adzm ◴[11 Apr 25 15:22 UTC] No.43654878[source]▶
Adobe is the one major company trying to be ethical with its AI training data and no one seems to even care. The AI features in Photoshop are the best around in my experience and come in handy constantly for all sorts of touchup work.

Anyway I don't really think they deserve a lot of the hate they get, but I do hope this encourages development of viable alternatives to their products. Photoshop is still pretty much peerless. Illustrator has a ton of competitors catching up. After Effects and Premiere for video editing are getting overtaken by Davinci Resolve -- though for motion graphics it is still hard to beat After Effects. Though I do love that Adobe simply uses JavaScript for its expression and scripting language.

replies(36): >>43654900 #>>43655311 #>>43655626 #>>43655700 #>>43655747 #>>43655859 #>>43655907 #>>43657271 #>>43657436 #>>43658069 #>>43658095 #>>43658187 #>>43658412 #>>43658496 #>>43658624 #>>43659012 #>>43659378 #>>43659401 #>>43659469 #>>43659478 #>>43659507 #>>43659546 #>>43659648 #>>43659715 #>>43659810 #>>43660283 #>>43661100 #>>43661103 #>>43661122 #>>43661755 #>>43664378 #>>43664554 #>>43665148 #>>43667578 #>>43674357 #>>43674455 #
numpad0 ◴[11 Apr 25 20:59 UTC] No.43658496[source]▶
What it implies is, it's not really about ethics per se, just like it's not really about 6th digits per se. People hate AI images, cut and dry.

Law is agreeable hate, in a way. Things that gets enough hate will get regulated out, sooner or later.

replies(3): >>43658963 #>>43659528 #>>43661593 #
TeMPOraL ◴[11 Apr 25 21:45 UTC] No.43658963[source]▶
> People hate AI images, cut and dry.

People hate bad AI images, because they hate bad images, period. They don't hate good AI images, and when they see great AI images, they don't even realize they are made by AI.

It's true, there's a deluge of bad art now, and it's almost entirely AI art. But it's not because AI models exist or how they're trained - it's because marketers[0] don't give a fuck about how people feel. AI art is cheap and takes little effort to get - it's so cheap and low-effort, that on the lower end of quality scale, there is no human competition. It makes no economic sense to commission human labor to make art this bad. But with AI, you can get it for free - and marketing loves this, because, again, they don't care about people or the commons[1], they just see an ability to get ahead by trading away quality for greater volume at lower costs.

In short: don't blame bad AI art on AI, blame it on people who spam us with it.

--

[0] - I don't mean here just marketing agencies and people with marketing-related job titles, but also generally people engaging in excessive promotion of their services, content, or themselves.

[1] - Such as population-level aesthetic sensibilities, or sanity.

replies(2): >>43664352 #>>43666527 #
gs17 ◴[12 Apr 25 17:51 UTC] No.43666527{3}[source]▶
> They don't hate good AI images, and when they see great AI images, they don't even realize they are made by AI.

There's a decent size group of people who have a knee-jerk negative response toward AI regardless of quality. They'd see that image, like it, and then when told it's AI, turn on it and decide it was obviously flawed from the beginning. Is there a version of "sour grapes" where the fox did eat the grapes, they were delicious, but he declared they were sour after the fact to claim moral superiority?

replies(2): >>43669256 #>>43669668 #
numpad0 ◴[13 Apr 25 02:49 UTC] No.43669668{4}[source]▶

  > and then when told it's AI, turn on it and decide it was obviously flawed from the beginning.
Have you seen any experimental results from researches in which participants were _falsely_ told something was AI-made, to prove and gauge that "moral superiority" effect? I'm not aware of any. There has to be many, because it has to be easy. No?
replies(1): >>43673942 #
1. gs17 ◴[13 Apr 25 16:28 UTC] No.43673942{5}[source]▶
https://www.nature.com/articles/s41598-023-45202-3 is pretty similar to that, they randomized AI/human-made labels and participants considered the exact same piece less valuable and less creative when labeled as AI-made. It's not measuring "moral superiority", but it shows a "negative response toward AI regardless of quality". It's definitely an irrational response.
replies(1): >>43708698 #
2. numpad0 ◴[16 Apr 25 18:22 UTC] No.43708698[source]▶
yeah that's measuring effects of labeling, not discrepancies between human made and AI generated.