News China will enforce clear flagging of all AI generated content starting from September

[Admin]

The Cyberspace Administration of China just released regulations that requires the explicit marking of AI-generated content — both visually and in its metadata.

China will enforce clear flagging of all AI generated content starting from September : Read more

 

[punkncat]

In my very limited experience with and understanding of AI content, I see this as a positive move. The ability to create such lifelike "video" of people, places, events can make it quite difficult to know if what was produced is the real thing. Basically, hope this resolves issues related to something like the beginning of Running Man (movie) to become reality.

 

[rluker5]

I agree with punkncat. AI can be entertaining and that entertainment is not significantly depreciated by identifying it. But conclusions drawn from AI in text are suspect as they generally are built just to sound good, not to be accurate. And AI is good at making fakes that aren't directly drawn from observation of reality but, unlike a cartoon, seem like they are. They already have notices for dramatic reenactments and AI constructs should follow similar rules.

People should know.

 

[hotaru251]

I see this as a positive move.

it is.
All ai isn't bad but it should clearly be stated to be using ai just so those who do care can know.

 

[bit_user]

Nevertheless, this move will help reduce misinformation from anyone and everyone, especially as AI LLMs become more advanced. By ensuring that artificially generated content is marked clearly, people could more easily determine if they’re looking at or listening to a real event or something conjured by a machine on some server farm.

You're assuming it will be uniformly enforced. Selective enforcement of such a law - including false claims by the state that real content is just unlabelled AI-generated content - could be used as both a way to support false narratives which support state objectives and suppress even true evidence backing narratives which are unfavorable.

As uneasy as I am about the potential of AI-generated content to misinform, I'm also very uneasy about the potential of such rules - especially without a fair, proportionate, and transparent judicial process.

 

[AkroZ]

One interest of this law is to be able to ignore AI content for the training of a new AI model.

 

[rluker5]

Speaking of AI, is anyone else using Edge and getting the offer for complete monitoring in exchange for "Copilot Vision"? It gets to "see what you see" and responds to your voice. It is popping up on my B580, 13600k living room pc and my 3080, 13900kf gaming pc.

I don't want anything to do with it, but would like to continue using Edge and think that it is only a matter of time before I will have to choose between the two.

 

[usertests]

I don't want anything to do with it, but would like to continue using Edge and think that it is only a matter of time before I will have to choose between the two.

Switch to Brave.

But if you don't want to, try finding these settings and see if they can turn it off or at least the popup/offer:

How to Enable or Disable Copilot Vision in Microsoft Edge • AppDeployNews

Copilot Vision is a feature within the Microsoft Edge browser that enhances user interaction with web content through AI-powered tools. It allows users to ask complex questions, get summarized information, and perform various actions directly from the browser sidebar. Here are some of the...

www.appdeploynews.com

If that doesn't work, maybe you can uninstall something to cripple it OS-wide.

 

[Mr Majestyk]

And Trump at fElon's and Altman's request is wanting to getting rid of all safeguards on AI.

 

[wwenze1]

This is nice. But what about photoshop pictures and video filters.

 

[TechieTwo]

After all the hacking from the CCP I wouldn't hold out a lot of hope that these rules will be effective. Unless 51 CIA agents wrote a letter confirming it to be Russian disinformation how can the average person know what is real and what is state sponsored propaganda?

 

[ThereAndBackAgain]

In my very limited experience with and understanding of AI content, I see this as a positive move. The ability to create such lifelike "video" of people, places, events can make it quite difficult to know if what was produced is the real thing.

I have wondered whether the advent of AI fakes will eventually result in video evidence being inadmissible in court, or at least regarded as suspect due to the potential of fakery.

Speaking of AI, is anyone else using Edge and getting the offer for complete monitoring in exchange for "Copilot Vision"? It gets to "see what you see" and responds to your voice. It is popping up on my B580, 13600k living room pc and my 3080, 13900kf gaming pc.

I don't want anything to do with it, but would like to continue using Edge and think that it is only a matter of time before I will have to choose between the two.

I have not seen that. But I set a couple of keys in the registry that controls Edge group policy to totally disable the sidebar and related elements, like, I believe, Copilot.

Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Edge

HubsSidebarEnabled = 0
StandaloneHubsSidebarEnabled = 0

Edit for clarity: These should be DWORD values.

 

[usertests]

I disagree with pukenecat, I prefer what will actually happen, spotty or non-existent enforcement with no watermarking. Screw your reality, rules, and yo lives, full steam ahead, LET'S GOOOOOOOO

(not sarcasm)

 

[rluker5]

Switch to Brave.

But if you don't want to, try finding these settings and see if they can turn it off or at least the popup/offer:

How to Enable or Disable Copilot Vision in Microsoft Edge • AppDeployNews

Copilot Vision is a feature within the Microsoft Edge browser that enhances user interaction with web content through AI-powered tools. It allows users to ask complex questions, get summarized information, and perform various actions directly from the browser sidebar. Here are some of the...

www.appdeploynews.com

If that doesn't work, maybe you can uninstall something to cripple it OS-wide.

I'll keep Brave in mind. I've gotten used to the sharing of browser data between my house PCs and phone though and wouldn't try that with Google. But the autofill does get to be a real nuisance sometimes so I will be glad to be rid of that. Also the guide must be outdated or the copilot vision implementation is half baked, but maybe since I haven't yet given permission, the copilot options just start copilot trying to be "helpful" in other ways.

I have wondered whether the advent of AI fakes will eventually result in video evidence being inadmissible in court, or at least regarded as suspect due to the potential of fakery.


I have not seen that. But I set a couple of keys in the registry that controls Edge group policy to totally disable the sidebar and related elements, like, I believe, Copilot.

Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Edge

HubsSidebarEnabled = 0
StandaloneHubsSidebarEnabled = 0

Thanks. I'll give that a try on my next pc that is just waiting to give me the copilot vision spiel in my office before I open up Edge -tomorrow. Then see what it does with the others.

 

[bit_user]

I have wondered whether the advent of AI fakes will eventually result in video evidence being inadmissible in court, or at least regarded as suspect due to the potential of fakery.

I've read about video attestation technologies for many years, which purport to establish end-to-end authenticity of video as what was actually recorded by a camera. I think it does depend on vendor-supplied tools to check the signatures, which means that if a security camera vendor was hacked or pressured by a government, that 3rd party would be able to forge the signature on generated or modified footage, in order to make it authenticate.

From what I've heard, the main thing about video evidence in court is that you need a human willing to testify under oath that it accurately reflects what happened. So, that's why I think such attestation isn't a bigger deal. However, we've learned a lot about how malleable human memory is, so I think that's not such a water-tight safeguard.

 

[edzieba]

You're assuming it will be uniformly enforced. Selective enforcement of such a law - including false claims by the state that real content is just unlabelled AI-generated content - could be used as both a way to support false narratives which support state objectives and suppress even true evidence backing narratives which are unfavorable.

As uneasy as I am about the potential of AI-generated content to misinform, I'm also very uneasy about the potential of such rules - especially without a fair, proportionate, and transparent judicial process.

The CCP can already do all of that, and have done so for quite some time. Through a combination of the 'great firewall' (filtering of wider internet content) and direct filtering of local networks, certain media, text phrases, etc, can all be filtered without any sort of fiction of slapping an "AI" label on it.
If anything, Chinese netizens will immediately attempt to exploit this by slapping an "AI generated" label on real subversive content for distribution as part of the ongoing and creative euphemism whackamole.

 

[George³]

And Trump at fElon's and Altman's request is wanting to getting rid of all safeguards on AI.

They work for you since this will wonderfully complement Order 13985. Then why do I feel discontent in your statement?

 

[ThereAndBackAgain]

Speaking of AI, is anyone else using Edge and getting the offer for complete monitoring in exchange for "Copilot Vision"? It gets to "see what you see" and responds to your voice. It is popping up on my B580, 13600k living room pc and my 3080, 13900kf gaming pc.

I don't want anything to do with it, but would like to continue using Edge and think that it is only a matter of time before I will have to choose between the two.

So, is this referring to Edge automatically opening a new tab saying "Congratulations, you now have access to Copilot Vision"? If so, I just had this happen to me. I just closed the tab. Hopefully that's the end of it. I don't think they're going to force anyone to use it who doesn't want to.

 

[rluker5]

So, is this referring to Edge automatically opening a new tab saying "Congratulations, you now have access to Copilot Vision"? If so, I just had this happen to me. I just closed the tab. Hopefully that's the end of it. I don't think they're going to force anyone to use it who doesn't want to.

Yes it is, and I don't agree. I think it is a matter of when, not if MS will enable the gift of Copilot Vision.

But I also closed the tab, after trying to find out through the tab how to disable it and it wanted me to opt in to the program before it would answer how to avoid it.

 

[ThereAndBackAgain]

Yes it is, and I don't agree. I think it is a matter of when, not if MS will enable the gift of Copilot Vision.

But I also closed the tab, after trying to find out through the tab how to disable it and it wanted me to opt in to the program before it would answer how to avoid it.

The day they make it mandatory with no means of disabling it is the day they permanently lose me as an Edge user. Which would be unfortunate, since I like the vertical tabs.

 

[OkinSama]

All a regulation like this will accomplish, is to create a false sense of security among the public, that they can trust things that aren't labeled as AI. People who aren't interested in deceiving others, will abide by the regulation.. Those who want to deceive people, won't.