8 ms·

Anyone feel like with the flood of AI generated content there's a risk of the past being 'erased'. Like in 10 years we won't be able to tell if any information

Anyone feel like with the flood of AI generated content there's a risk of the past being 'erased'. Like in 10 years we won't be able to tell if any information from the past is real or fake - sounds, pictures, videos, etc.. Like we need to start cryptographically signing all content now if there's any hope of being able to verify it as 'real' 10 years from now.

seydor 3y ago
The past ended in 2022
- shon 3y ago
  This ^^
- apabepa 3y ago
  Or is the past endlessly rehashed with AI generated content?
- bottlepalm 3y ago
  Agree. Any video/image/text created post-2022 is now suspect of being AI generated (even this comment). And without any 'registering' of pre-2022 content, we can easily lose track and not really know what from pre-2022 is authentic or not. Maybe it's not a big deal to 'lose' the past, maybe landfills will be mined for authentic content.
  - seydor 3y ago
    There should be a canonical copy of the 2022 internet that is verifiable. Archive.org is not enough
jeffwass 3y ago
I’ve been wondering about this and real video evidence (eg dashcam or cctv) being refuted in court for inability to show it’s not deepfaked.
crazygringo 3y ago
No. We've had photo and audio manipulation for many decades now. For a long time now, we've had to separate out what's credible from what's bullshit. Fortunately, it's pretty simple in real life. We have certain publications and sources we trust, whether they're the NYT or a respected industry blog. We know they take accurate reporting seriously, fire journalists who are caught fabricating things, etc. If we see a clip on YouTube from the BBC, we can trust it's almost certainly legit. If it's some crazy claim from a rando and you care whether it's real, it's easy to look it up to see if anyone credible has confirmed it. So no, no worry at all about the past being erased.
- minsc_and_boo 3y ago
  Yep, every time technology shifts, reputation systems shift in response. This goes all the way back to yellow news with newspapers: https://en.wikipedia.org/wiki/Yellow_journalism https://en.wikipedia.org/wiki/Yellow_journalism
- ysleepy 3y ago
  I don't agree. With ML tools it is possible to make sweeping changes to images and text that are often impossible to detect. combined with the centralisation of most online activities, large players could alter the past. Imagine facebook decides to subtly change every public post and comment to show some particular person or cause in a better light.
  - crazygringo 3y ago
    If one "large player" like the NYT decides to "alter the past", you can compare with the WaPo or any other newspaper. You can compare with the Internet Archive. You can compare with microfiche. These aren't "impossible to detect", they're trivial to detect if you bother to compare. We have tons of credible archived sources owned by different institutions. And these sources are successful in large part due to their credibility and trustworthiness. It's just not economically rational for any of them to start "altering the past", and if they did, they'd be caught basically immediately and their reputation would be ruined. This isn't an ML/tooling question, it's a question of humans and reputation and economic incentives.
randcraw 3y ago
With 90% of human generated media content being forgettable within weeks of publication, and AI not yet capable of matching even average human content (much less pro level), it’ll be some time before we have to worry about AI overwhelming most media content and erasing the works of memorable human authors.
- famouswaffles 3y ago
  >and AI not yet capable of matching even average human content (much less pro level) Yeah this is not true. Sota Text, Image generation is well above average baselines. You can certainly generate professional level art on Midjourney
  - squidsoup 3y ago
    Commercial art and Art are not the same thing.
    - bottlepalm 3y ago
      Unless you have a tool that can tell the difference, they are.
JohnFen 3y ago
Yes, this is one of my concerns about all of this. The danger is real.
russdill 3y ago
If you're watching a movie or TV show, a vast majority of the sounds you are hearing are not "real". Has that bothered you before?
- swores 3y ago
  That seems as pointless a question as suggesting that enjoying TV shows means you shouldn't care if everyone in your life constantly lies to you.
- Ylpertnodi 3y ago
  >the stuff i hear is real. Perhaps you meant 'are not from the actual source you think they are'? *my favorite is always the nightclub scene that goes real quiet when the actors act using their voices (which are real, but may be dubbed in afterwards).
vagab0nd 3y ago
Even with digital signatures, there are limits to what we can really verify. We'll likely be able to verify whether an entity is a real human, using some kind of "proof of humanity" system. We will have cameras/mics with private keys built-in. The content can be signed as it's produced. But in this case, what's stopping me from recording a fake recording? Maybe it's a non-issue. We used text to record history and we've been able to manipulate that since, well, forever.

seydor 3y ago

The past ended in 2022

shon 3y ago
This ^^
apabepa 3y ago
Or is the past endlessly rehashed with AI generated content?
bottlepalm 3y ago
Agree. Any video/image/text created post-2022 is now suspect of being AI generated (even this comment). And without any 'registering' of pre-2022 content, we can easily lose track and not really know what from pre-2022 is authentic or not. Maybe it's not a big deal to 'lose' the past, maybe landfills will be mined for authentic content.
- seydor 3y ago
  There should be a canonical copy of the 2022 internet that is verifiable. Archive.org is not enough

jeffwass 3y ago

I’ve been wondering about this and real video evidence (eg dashcam or cctv) being refuted in court for inability to show it’s not deepfaked.

crazygringo 3y ago

No. We've had photo and audio manipulation for many decades now. For a long time now, we've had to separate out what's credible from what's bullshit. Fortunately, it's pretty simple in real life. We have certain publications and sources we trust, whether they're the NYT or a respected industry blog. We know they take accurate reporting seriously, fire journalists who are caught fabricating things, etc. If we see a clip on YouTube from the BBC, we can trust it's almost certainly legit. If it's some crazy claim from a rando and you care whether it's real, it's easy to look it up to see if anyone credible has confirmed it. So no, no worry at all about the past being erased.

minsc_and_boo 3y ago
Yep, every time technology shifts, reputation systems shift in response. This goes all the way back to yellow news with newspapers: https://en.wikipedia.org/wiki/Yellow_journalism https://en.wikipedia.org/wiki/Yellow_journalism
ysleepy 3y ago
I don't agree. With ML tools it is possible to make sweeping changes to images and text that are often impossible to detect. combined with the centralisation of most online activities, large players could alter the past. Imagine facebook decides to subtly change every public post and comment to show some particular person or cause in a better light.
- crazygringo 3y ago
  If one "large player" like the NYT decides to "alter the past", you can compare with the WaPo or any other newspaper. You can compare with the Internet Archive. You can compare with microfiche. These aren't "impossible to detect", they're trivial to detect if you bother to compare. We have tons of credible archived sources owned by different institutions. And these sources are successful in large part due to their credibility and trustworthiness. It's just not economically rational for any of them to start "altering the past", and if they did, they'd be caught basically immediately and their reputation would be ruined. This isn't an ML/tooling question, it's a question of humans and reputation and economic incentives.

randcraw 3y ago

With 90% of human generated media content being forgettable within weeks of publication, and AI not yet capable of matching even average human content (much less pro level), it’ll be some time before we have to worry about AI overwhelming most media content and erasing the works of memorable human authors.

famouswaffles 3y ago
>and AI not yet capable of matching even average human content (much less pro level) Yeah this is not true. Sota Text, Image generation is well above average baselines. You can certainly generate professional level art on Midjourney
- squidsoup 3y ago
  Commercial art and Art are not the same thing.
  - bottlepalm 3y ago
    Unless you have a tool that can tell the difference, they are.

JohnFen 3y ago

Yes, this is one of my concerns about all of this. The danger is real.

russdill 3y ago

If you're watching a movie or TV show, a vast majority of the sounds you are hearing are not "real". Has that bothered you before?

swores 3y ago
That seems as pointless a question as suggesting that enjoying TV shows means you shouldn't care if everyone in your life constantly lies to you.
Ylpertnodi 3y ago
>the stuff i hear is real. Perhaps you meant 'are not from the actual source you think they are'? *my favorite is always the nightclub scene that goes real quiet when the actors act using their voices (which are real, but may be dubbed in afterwards).

vagab0nd 3y ago

Even with digital signatures, there are limits to what we can really verify. We'll likely be able to verify whether an entity is a real human, using some kind of "proof of humanity" system. We will have cameras/mics with private keys built-in. The content can be signed as it's produced. But in this case, what's stopping me from recording a fake recording? Maybe it's a non-issue. We used text to record history and we've been able to manipulate that since, well, forever.