Live data from Hacker News

YouTube now requires to label their realistic-looking videos made using AI

blog.google

451–460 of 517 posts

Re: YouTube now requires to label their realistic-looking videos made using AI

#451
post #37

I think it's smart to start trying things here. This has infinite flaws with it, but from a business and learnings standpoint it's a step toward the right direction. Over time we're going to both learn and decide what is and isn't important to designate as "AI" - Google's approach here at least breaks this into rules of what "AI" things are important to label: • Makes a real person appear to say or do something they…

Basically, Google decides what’s real and what’s not. Cool.

At least on Youtube, Google's AI does, yes.

And? Do they have a track record of crafting a false reality for people?

Re: YouTube now requires to label their realistic-looking videos made using AI

#452
post #37

I think it's smart to start trying things here. This has infinite flaws with it, but from a business and learnings standpoint it's a step toward the right direction. Over time we're going to both learn and decide what is and isn't important to designate as "AI" - Google's approach here at least breaks this into rules of what "AI" things are important to label: • Makes a real person appear to say or do something they…

But why does it matter if it was generated by AI instead of generated like Forest Gump?

Re: YouTube now requires to label their realistic-looking videos made using AI

#453
post #37

I think it's smart to start trying things here. This has infinite flaws with it, but from a business and learnings standpoint it's a step toward the right direction. Over time we're going to both learn and decide what is and isn't important to designate as "AI" - Google's approach here at least breaks this into rules of what "AI" things are important to label: • Makes a real person appear to say or do something they…

> I am curious to see the legal arguments that will inevitably kick up from each of these

Google policy isn't law; there's no court judging legal arguments, it is enforced at Google’s whim and with effectively no recourse, at least not one which is focused on parsing arguments about the details of the policy.

So there won’t be “legal arguments” over what exactly it applies to.

Re: YouTube now requires to label their realistic-looking videos made using AI

#454

This is interesting because it highlights the trust we've always placed in real-looking images. It brings real-looking images down to the same level as text. It's always been possible to write fake news. We've never had to add disclaimers at the top of textual content, e.g. "This text is made to sound like it describes real events, but contains invented and/or inaccurate facts." We feel the need to add this to video…

>We've never had to add disclaimers at the top of textual content, e.g. "This text is made to sound like it describes real events, but contains invented and/or inaccurate facts."

I mean, yes we have? https://en.wikipedia.org/wiki/All_persons_fictitious_disclai...

Re: YouTube now requires to label their realistic-looking videos made using AI

#455
post #37

I think it's smart to start trying things here. This has infinite flaws with it, but from a business and learnings standpoint it's a step toward the right direction. Over time we're going to both learn and decide what is and isn't important to designate as "AI" - Google's approach here at least breaks this into rules of what "AI" things are important to label: • Makes a real person appear to say or do something they…

But why does it matter if it was generated by AI instead of generated like Forest Gump?

Because right now AI is an issue the public and policymakers are concerned about, and this is to show that private industry can take adequate steps to control it to stave up government regulation while the attention is high.

Re: YouTube now requires to label their realistic-looking videos made using AI

#456

Earlier quoted context omitted.

Porn classification / regulation boils down to: "I'll know it when I see it." Implying the existence of some hyper vigilant seer who can heroically determine what we should keep behind the video storre curtain of dencey, as if no grey areas exist. This also has the problem of requiring actual unbiased humans to view and accurately assess everything, which of course does not scale. Perhaps AI classification is the mir…

The great thing about AI, is that it's exactly optimized for discriminating in gray areas with difficult-to-articulate rules like "I'll know it when I see it."

You mean it is exactly optimized to extend our conscious and unconscious biases to gray areas in an inconsistent and arbitrary way?

Re: YouTube now requires to label their realistic-looking videos made using AI

#457
post #213

Earlier quoted context omitted.

The AI tags are fundamentally useless. The premise is that it would prevent someone from misleading you by thinking that something happened when it didn't, but someone who wants to do that would just not tag it then. Which is where the real abuse comes in: You post footage of a real event and they say it was AI, and ban you for it etc., because what actually happened is politically inconvenient. And the only way to p…

I think you have a bit backwards. If you want to publish pixels on a screen there should be no assumption that they represent real events. If you want to publish proof of an event, you should have some pixels on a screen along with some cryptographic signature from a device sensor that would necessitate atleast a big corporation like Nikon / Sony / etc. being "in on it" to fake. Also since no one likes RAW footage it…

> If you want to publish proof of an event, you should have some pixels on a screen along with some cryptographic signature from a device sensor that would necessitate atleast a big corporation like Nikon / Sony / etc. being "in on it" to fake.

That idea doesn't work, at all.

Even assuming a perfect technical implementation, all you'd have to do to defeat it is launder your fake image through a camera's image sensor. And there's even a term for doing that: telecine.

With the right jig, a HiDPI display, and typical photo editing (no one shows you raw, full-res images), I don't think such a signature forgery would detectable by a layman or maybe even an expert.

Re: YouTube now requires to label their realistic-looking videos made using AI

#458
post #390

Earlier quoted context omitted.

> You share both the processed version and the original, as proof that there's no misleading alteration so you cannot share the original if you intend to black out something from the original that you don't want revealed (e.g., a face or name or something). The way you specced out how a signed jpeg works means the raw data _must_ remain visible. There's gonna be unintended consequences from such a system. And it aint…

> The way you specced out how a signed jpeg works means the raw data _must_ remain visible. There's gonna be unintended consequences from such a system. You can obviously extend this if you want to add bells and whistles like cropping or whatever. Like signing every NxN sub-block separately, or more fancy stuff if you really care. It should be obvious I'm not going to design in every feature you could possibly dream…

It also occurs to me that the camera chips -- or even separately-sold chips -- could be augmented to perform transformations (like black-out) on already-signed images. You could even make this work with arbitrary transformations - just sign the new image along with a description (e.g., bytecode) of the sequence of transformations applied to it so far. This would let you post-process authentic images while maintaining authenticity.

The possibilities are pretty endless here.

Re: YouTube now requires to label their realistic-looking videos made using AI

#459
post #45

Earlier quoted context omitted.

One black hat thing I'm curious about though is whether or not this tag can be weaponized. If I upload a real event and tag it as AI, will it reduce user trust that the real event ever happened?

The AI tags are fundamentally useless. The premise is that it would prevent someone from misleading you by thinking that something happened when it didn't, but someone who wants to do that would just not tag it then. Which is where the real abuse comes in: You post footage of a real event and they say it was AI, and ban you for it etc., because what actually happened is politically inconvenient. And the only way to p…

They’re just gathering training data to train their AI-detection models.

Re: YouTube now requires to label their realistic-looking videos made using AI

#460

Earlier quoted context omitted.

I see a lot of confusing authenticity with accuracy. Someone can sign the statement "Obama is white" but that doesn't make it a true statement. The use of PKI as part of showing provenance/chain of trust doesn't make any claims about the accuracy of what is signed. All it does is assert that a given identity signed something.

It's not about what is being signed, it's about who signed it and whether you trust that source. I want credible news outlets to start signing their content with a key I can verify as theirs. In that future all unsigned content is by definition fishy. PKI is the only way to implement trust in a digital realm.

Why do you need a whole PKI for that, rather than just, say, a link to the news outlet's website where the content is hosted? People have already been doing that pretty much since the web was created.
Post reply on HN