4 ms·

> It's still a computer, and we still shouldn't trust everything we see. The fundamentals haven't changed. I think that by now it should be crystal clear to e

> It's still a computer, and we still shouldn't trust everything we see. The fundamentals haven't changed. I think that by now it should be crystal clear to everyone that it matters a lot the sheer scale a new technology permits for $nefarious_intent. Knives (under a certain size) are not regulated. Guns are regulated in most countries. Atomic bombs are definitely regulated. They can all kill people if used badly, though. When a photo was faked/composed with old tech, it was relatively easy to spot. With photoshop, it became more complicated to spot it but at the same time it wasn't easy to mass-produce altered images. Large models are changing the rules here as well.

hk__2 1y ago
> Knives (under a certain size) are not regulated. Guns are regulated in most countries. Atomic bombs are definitely regulated I don’t think this is a good comparison: knives are easy to produce, guns a bit harder, atomic bombs definitely harder. You should find something that is as easy to produce as a knife, but regulated.
- darkwater 1y ago
  The "product" to be regulated here is the LLM/model itself, not its output. Or, if you see the altered photo as the "product", then the "product" of the knife/gun/bomb is the damage it creates to a human body.
- wing-_-nuts 1y ago
  >You should find something that is as easy to produce as a knife, but regulated. The DEA and ATF have entered the chat
  - withinboredom 1y ago
    They can leave, plain water fits this bill.
csallen 1y ago
I think we're overreacting. Digital fakes will proliferate, and we'll freak out bc it's new. But after a certain amount of time, we'll just get used to it and realize that the world goes on, and whatever major adverse effects actually aren't that difficult to deal with. Which is not the case with nuclear proliferation or things like that. The story of human history is newer generations freaking about progress and novel changes that have never been seen before. And later generations being perfectly okay with it and adapting to a new style of life.
- darkwater 1y ago
  In general I concur but the adaptation doesn't come out of the blue or just only because people get used to it but also because countermeasures are taken, regulations are written and adjustments are made to reduce the negative impact. Also the hyperconnected society is still relatively new and I'm not sure we have adapted for it yet.
  - Yokohiii 1y ago
    Photography and motion pictures were deemed evil. Video games made you a mass murderer. Barcodes somehow seem to affect your health or the freshness of vegetables. The earth is flat. The issue is that some people believe shit someone tells them and they deny any facts. This has been always a problem. I am all in for labeling content as AI generated. But it wont help with people trying to be malicious or who choose to be dumb. Forcing to watermark every picture made neither, it will turn into a massive problem, its a solid pillar towards full scale surveillance. Just alone the fact that analog cams become by default less trustworthy then any digital device with watermarking is terrible. Even worse, phones will eventually have AI upscaling and similar by default, you can't even make an accurate picture without anything being tagged AI. The information is eventually worthless.
- SV_BubbleTime 1y ago
  It shouldn’t be that we panic about it and regulate the hell out. We could use the opportunity to deploy robust systems of verification and validation to all digital works. One that allows for proving authenticity while respecting privacy if desired. For example… it’s insane in the US we revolve around a paper social security number that we know damn well isn’t unique. Or that it’s a massive pain in the ass for most people to even check the hash of a download. Guess which we’ll do!
commandlinefan 1y ago
> a new technology permits for $nefarious_intent But people with actual nefarious intent will easily be able to remove these watermarks, however they're implemented. This is copy protection and key escrow all over again - it hurts honest people and doesn't even slow down bad people.

hk__2 1y ago

> Knives (under a certain size) are not regulated. Guns are regulated in most countries. Atomic bombs are definitely regulated I don’t think this is a good comparison: knives are easy to produce, guns a bit harder, atomic bombs definitely harder. You should find something that is as easy to produce as a knife, but regulated.

darkwater 1y ago
The "product" to be regulated here is the LLM/model itself, not its output. Or, if you see the altered photo as the "product", then the "product" of the knife/gun/bomb is the damage it creates to a human body.
wing-_-nuts 1y ago
>You should find something that is as easy to produce as a knife, but regulated. The DEA and ATF have entered the chat
- withinboredom 1y ago
  They can leave, plain water fits this bill.

csallen 1y ago

I think we're overreacting. Digital fakes will proliferate, and we'll freak out bc it's new. But after a certain amount of time, we'll just get used to it and realize that the world goes on, and whatever major adverse effects actually aren't that difficult to deal with. Which is not the case with nuclear proliferation or things like that. The story of human history is newer generations freaking about progress and novel changes that have never been seen before. And later generations being perfectly okay with it and adapting to a new style of life.

darkwater 1y ago
In general I concur but the adaptation doesn't come out of the blue or just only because people get used to it but also because countermeasures are taken, regulations are written and adjustments are made to reduce the negative impact. Also the hyperconnected society is still relatively new and I'm not sure we have adapted for it yet.
- Yokohiii 1y ago
  Photography and motion pictures were deemed evil. Video games made you a mass murderer. Barcodes somehow seem to affect your health or the freshness of vegetables. The earth is flat. The issue is that some people believe shit someone tells them and they deny any facts. This has been always a problem. I am all in for labeling content as AI generated. But it wont help with people trying to be malicious or who choose to be dumb. Forcing to watermark every picture made neither, it will turn into a massive problem, its a solid pillar towards full scale surveillance. Just alone the fact that analog cams become by default less trustworthy then any digital device with watermarking is terrible. Even worse, phones will eventually have AI upscaling and similar by default, you can't even make an accurate picture without anything being tagged AI. The information is eventually worthless.
SV_BubbleTime 1y ago
It shouldn’t be that we panic about it and regulate the hell out. We could use the opportunity to deploy robust systems of verification and validation to all digital works. One that allows for proving authenticity while respecting privacy if desired. For example… it’s insane in the US we revolve around a paper social security number that we know damn well isn’t unique. Or that it’s a massive pain in the ass for most people to even check the hash of a download. Guess which we’ll do!

commandlinefan 1y ago

> a new technology permits for $nefarious_intent But people with actual nefarious intent will easily be able to remove these watermarks, however they're implemented. This is copy protection and key escrow all over again - it hurts honest people and doesn't even slow down bad people.