Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)B

brucethemoose

@ brucethemoose @lemmy.world

Posts
18
Comments
2663
Joined
2 yr. ago

  • Android users can't remove nude-scanning app as Google opens the feature to app developers

    Jump

  • Okay, thinking on this more, I have a very hot take here.

    What's the big deal?

    Google is seemingly offering a locally processed machine vision API to other apps, as far as I tell. This might manifest as some messaging service nagging you if you try to send nudes to someone, or filtering porn out of a feed. Or other "safety" features I suppose, like classifying slur-laden messages as abusive.

    ...But Google was most certainly doing this with Google apps already. Like the stock photo library most apps use.

    If people are already using unprivate services for this kind of thing, honestly, they were digging their own grave, and this hardly moves the needle. All it does is make the spying/censorship more apparent.

    I say those users need a cold bucket of water to the face, and this may provide it.

    ...But otherwise, it wont affect apps that dont use the API, right?

  • Android users can't remove nude-scanning app as Google opens the feature to app developers

    Jump

  • You may come off as hyperbolic, but this is exactly what Canon is trying to do.

    They already try to funnel users into a canon.cloud subscription with cloud editing features, and a DPP editing software subscription. Their HDR mode uses a weird curve that makes using anything but their software annoying. They made their crappy free webcam app subscription only, and neutered many cameras so that's the only option to use.

    Not to speak of locking the RF mount out of third parties...

    But to be fair, Canon is only doing that because they are #1 in marketshare, though. Panasonic, Fuji, OM aren't behaving so badly. And the Chinese pocket cam makers are rapidly catching up (though obviously they have their own enshittification issues).

  • Android users can't remove nude-scanning app as Google opens the feature to app developers

    Jump

  • Old "pro" cameras like full-frame SLRs are bulky though. That's not always ergonomic to have.

    And honestly, modern phones/gadgets with a lofic sensor (like Xiaomi or Hisense now, Apple soon) are absolutely crazy. They blow the dynamic range and shadow noise of old full frame sensors out of the water, and give new ones a run for their money:

    https://www.cined.com/dji-osmo-pocket-4p-lab-test-rolling-shutter-dynamic-range-and-exposure-latitude/

    Don't get me wrong; you are preaching to the choir here. I love cameras. I am trading in my Canon and buying a Fuji or Panasonic ILC soon.

    But I think its a mistake to underestimate modern smartphones based on how bad older phones were. The sensors have gotten big, their sensor tech is a few years ahead of dedicated cameras, and you can bypass the overprocessed look with 3rd party apps. That, and lofic is the biggest jump in sensor tech in a decade+, and its only coming to small sensor cameras for now.

  • Android users can't remove nude-scanning app as Google opens the feature to app developers

    Jump

  • Apple already uses machine vision for photo searching, that weird auto categorization it does and such. They have for years.

    I would bet the photos app already classifies nudes, but silenty, mostly so it doesnt add your nudes to an auto album.

  • Android users can't remove nude-scanning app as Google opens the feature to app developers

    Jump

  • Eh, they are alright. Issue is they dont get better than smartphones until the cameras are expensive.

  • Android users can't remove nude-scanning app as Google opens the feature to app developers

    Jump

  • Those kinds of cameras didnt disappear.

    But Canon already enshittifies their bodies some, with links to the canon.cloud subscriptions in-camera (which is a front for RAW training if you read the ToS), a subscription for linked editing software, and some feature crippling to push them.

    There is actually healthy competition in the space, though.

  • Do not use ai

    Jump

  • Fighting local models is playing right into enshittification, though.

    They WANT people to be scared of "AI." They want people to keep talking about it as this existential anthropomorphized danger, not a tool theyve enshittified.

    Because the thing that keeps Altman/Modi up at night is the proliferation of trifty, dirt cheap open weights models. Ones presented as tools, not intelligence. And every bit people rail against them is more ammunition to regulate those open models away, so they have a monopoly.

  • AI/LLM Usage Becoming A "Denial of Service Attack" On Open-Source Project Maintainers

    Jump

  • In my experience, the users behind this tend to be a state of AI Psychosis.

    They're the kind of user who forked or started some giant project, with a wall of text and emojiis and tables in the readme, yet theres no concrete information on how well it actually works. They submit some gigantic PR that touches a ton of files... for a feature that already exists in the project, and they apparently werent aware of.

    And this is another tell. This is not some rogue agent like people think, its some person deep down a rabbit hole that might think they're contributing to QEMU. They arent necessarily malicious; just a bit crazy.

    Malicious versions do exist, but thats usually motivated by grifters trying to pad their resume with contributions. Theyd do that on GitHub, not Gitlab.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • I dont think any spitting is necessary, nor is it productive.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • Some models do use open datasets and largely reproducible training, with the Nvidia Nemotron series being the highest profile example. You are right that many are opaque about training data, though, and this is not okay.

    Using them is just as unethical as using the models from OpenAI.

    And I vehemently disagree with this.

    OpenAI is a different order of magnitude of moral depravity. Thats like saying using Lemmy is just as bad as using Facebook or Palantir.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • That is called luma noise, yes. Heh. Its nothing new.

    The thing that I wish was better understood is that “slop” = “low effort.” ...that means that not all AI is slop, and thus if you really want to detect AI, you can’t rely on it always being slop.

    Yes! I would wear this on my forehead on Lemmy, if I could.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • I mean, the customers are the ones with the purchasing power here, and most of the shame was put on the studio.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • Also... you could just add grain via post processing?

    Its actually a reasonable technique, as it softens diffusion blocking artifacts the same way it softens video codec blocking artifacts.

    The YouTuber sorta kinda has a point, depending on the context. Spammers tend to use the absolute laziest lowest common denomenator, so its quite possible theyre largely using the same model, with the same artifacts, and not bothering to edit output. So maybe they did have a common noise pattern? Though the way thats presented still sounds fishy to me.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • You put this incredibly abrasively, but there is a grain of truth there. Gamers, in aggregate, repeatedly preorder or buy hyped garbage, so perhaps we dont have much of a leg to stand on when it comes to boycotts.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • Well it is actually complicated. It appears the tiny studio ran out of time/budget trying to get usable CGI, so they did some diffusion real-quick as an act of desperation, and got slop as a result.

    Not everyone is exactly up to speed on diffusion, either. I bet they Googled and used the first video diffusion service they found, and hence got bad results, but its also fair to criticize them for publishing them. They either didnt bother to proof the output, or they did and knowingly published anyway.

    Im not sure anyone is the villain here. But I dont like people leaning into outrage culture, and trying to paint it as black-and-white.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • Unfortunately it may be politicized beyond recovery, at least in the US with the trajectory of things.

    Its going to be associated with the corporations doing awful things with it, and thats going to blot out the sun.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • AI has a certain look to it

    I think this is a dangerous misconception.

    I could feed locally run Minimax H3 a set of reference art, like the painted stills from this game, and it would exactly match the style. I could use start/end frames, motion masks, pose control, character sheets, voice and music samples, reference videos; I can get it to do whatever I want with extreme precision, and as long as I checked the output it would be indistinguishable from CGI.

    The point I'm making is that a lot of people percieve diffusion as mashing a Stable Diffusion keyword prompt into a text box, and getting a sloppy result. Maybe thats still how shit corporate services like ChatGPT work.

    But a LOT has changed over two years.

    Theres no way people can definitively detect generative output anymore. We can IF its done lazily, but its quite possibly indistinguishable now, too.

  • A touch of AI is all it takes to trigger backlash, as a promising new indie falls afoul of slop skeptics

    Jump

  • Quality control, for one.

    Artifacts like this would be unnaceptable if they were from CGI or whatever, but somehow generative models get a pass? That makes no sense to me, yet it happens all the time; heavily distorted output somehow meets the bar for publishing.

    Honestly it has nothing to do with "AI." It would be worth calling out however it was made. But for some reason generative diffusion models always seem to lower the standards of output for user.

    If it was indistinguishable from CGI and fit the rest of the hand-painted style, the argument would be very different, IMO.

  • What's the most unforgivable sin done in media that you're still bitter about?

    Jump

  • Rare clearly shouldn’t be trusted with solo developing another game and neither should Bungie.

    I find the phrasing of this amusing. In my head, I'm picturing Arkane and Bungie in a room, and Sharma is pointing at Arkane and saying "Dont leave Bungie alone in a room with the cookie jar. Dont do it."