AI is really worrying and I fear this sort of image generation is tip of the iceberg

[mids2019]

https://www.bbc.co.uk/news/articles/c98p1r4e6m8o

I don't know to what extent current legislation covers this but to my mind any woman with an image on the net could be prone to this. Are we going to reach a stage where our daughters are going to simply not want any image taken of them for fear of how it could be manipulated?

The motivation is very clear, a quick look at the profiles of the users (which I have archived and sent to Ofcom if they wish to contact the police regarding these images) creating these images shows they are far-right Elon Musk supporters and seem very upset at the thought of the UK wanting to ban Twitter. The homophobic comments made towards the Ofcom staff are also vile and repulsive.

Yeah right 🙄

I think the answer involves -

(1) making it explicitly clear what is illegal (FWIW I would argue that any / all fake and real CSAM should be illegal, along with any non-consensual real porn and any non-consensual AI porn using real faces which an ordinary and reasonable person would see as having the potential to fool viewers into thinking it's real.) Punishing people who do illegal things very hard.

(2) making it explicitly clear that - unfortunately - if you go out in public, or share images of yourself with the public, then your face becomes public property and that you have lost all sorts of control. This does not give people the right to make fake videos of you claiming to support Nazis or make porn using your face, but it does gives an awful lot of people Article 10 rights which should only be restricted for very good reason. I do not need your consent to film you or photograph you as you walk down the street. I do not need your consent to download photos you publish of yourself, or I take of you when you are in public, and edit them to add a silly cartoon hitler moustache or to print out and use in a collage.

(3) teaching people resilience.

I think that your concerns are all valid, and we need to try very hard to ensure women (and men like me) are much more able to show our faces and real names online without fear of harm coming to us. BUT, other people have rights to and Article 10 rights of freedom of expression are vital.

If Elon Musk was really a great 3D chess player he could circumvent their masterplan by simply removing the taking women's/children's clothes of feature from Grok. Even if the images of Ofcom members were created by leftists using Grok in order to encourage a crackdown on X. removing the revenge form feature would immediately counter that. Checkmate woke moralists.

He won't, and they probably aren't. You are however correct that there are many creepy men on the left who would also quite happily use these features to target women on the right. And men of no political persuasion who will use it to target women and children just because. It's bad, no matter which side of the political spectrum you are. And indeed other AI models could also do this and I don't want this to become normalised.

I agree that people have rights to freedom of speech/expression.
If I add a silly Hitler moustache to your photo then that's something I have done as a human. It conceivably falls under freedom of expression/freedom of speech. If an AI programme generates an image of you dressed as Hitler then who conceivably has created the photo? The person who asked the AI to create a funny picture of you or the AI? I don't think algorithms are speech and I don't think computer programmes have the right to have their output protected in the same way people's speech is. And corporations aren't people.

So I think it's important actually to draw a distinction between "speech" produced by people (which can also include images, art, essays etc) and the output of programmes like From, Chat GPT. That doesn't mean ban it all. But it doesn't qualify for protection under free speech laws IMO.

I suspect that it isn't a trivial task to ensure Artificial "Intelligence" is unable to be used for this sort of purpose. The fact is, it isn't intelligent, so I think it would need a comprehensive set of rules. Who could draw up that set of rules so that no ingenious person could find a way to circumvent them? Certainly Musk should announce that his techies are working on this as a priority. He could also improve the frankly ineffective moderation. But I doubt if he has the ability to fix the problem completely in Grok, and I am sure the problem exists in other LLMs.

Free speech is just part of it, it is freedom of expression. From google AI.

Article 10 rights, primarily from the European Convention on Human Rights (ECHR) and the UK's Human Rights Act, guarantee freedom of expression, protecting your right to hold opinions and share information and ideas freely through various mediums (speech, art, media) without government interference, though this right is qualified and can be limited for specific, legitimate reasons like national security or protecting others' rights. This right covers political views, art, and even unpopular ideas, ensuring a free press and protecting journalists' sources.

As a society we need to try to work out where to draw the line, but from my point of view I believe that my rights to take photos or videos, or to create art using whatever primitive (paint and canvas) or sophisticated (AI) medium(s) / tools I wish should be incredibly wide, and only be prevented at points like malicious porn which could be seen as real, CSAM, libel etc.

It is very difficult to know the best way forward, but if there's one thing the last 5 years have taught me it is that freedom of expression and speech are two of the most important things we have, and they should be restricted as little as possible.

Musk could disable Grok image manipulation from the UK site but that doesn’t stop thousands of other AI programmes from manipulating images. Perhaps we should ban ALL image manipulation including filters? That would be revealing for a lot of trans-identified men!

What about this? I doubt any consent was involved…

https://www.bbc.co.uk/news/uk-england-london-64411569

Precisely, it would be outrageous if that satire was not allowed... and in my view (in some ways unfortunately) I believe that the moment you are in public giving an opinion you can be satirized, and if the public place you choose is social media with your face on it...

You could definitely use AI to create art. But I don't think asking Grok or any other.programme to draw me a picture of a bird is art.in the same way. If it is art/expression then it is as much the product of the original artists/photographers whose images of birds taught the machine to learn what bird is. Or it's a product of the people who wrote the code. But I dont think that it is me thats drawn the bird or is responsible for the picture of the bird existing in the form it is. The whole point of AI is that most of the decision making about the command "draw a bird" is happening inside a black box.

It sounds nitpicky but let's say the programmer of the AI hated robins and loved crows. They could skew their.AI to produce really negative images of robins stealing shit or beautiful crows. When I ask the AI to draw me a picture of a bird and it gives me a shifty Robin I might believe it's my own decision making that is predominantly responsible for the picture looking like that but I would be dangerously deluded. Yes, you could refine and refine the images to show exactly what you want. But overall AI gives people the illusion that they are more involved in the decisions for making AI images look how they look than they are. When actually a lot of it is completely hidden and in some cases completely unknowable.
Pen and paper won't do that to you. Neither will Adobe photoshop or computer drawing tools.

AI images are as much art as other art is. It may not reflect skill of the artist or be good but neither were my children’s art when they were toddlers. And manipulation of images produced by others has been considered art for as long as images were produced.

That was long, but basically you can end up thinking that in using AI you are expressing yourself, while in reality you are expressing the views of the person who owns the AI or the programmers who helped build it or the raw material that was fed into it. I.dont think it should automatically be regarded as expression in the same way even doodles or graffitied swear words are.

Art has to involve decisions. In the case of a famous artist maybe they choose to draw the flowers as large as the house to symbolise the importance of nature. In the case of a child's drawing that likely isn't a deliberate decision it's because they haven't learnt how to do perspective. But they have still made choices about where everything goes on the page. Likewise collages. I might be cutting pictures from a magazine but I'm deciding where to place them to make a point about sex and consumerism or whatever.
If you tell an AI to draw a house with flowers outside its not really an act of conscious self expression to the same extent. If I ask for a poem about beetles in the style of Sylvia Plath did I write the poem? Did Sylvia Plath? Did the machine?

Seriously? Thats how you see this? Not an attack on Ofcom using women as a proxy?

The UK will bring into force a law which will make it illegal to create non-consensual intimate images, following widespread concerns over Elon Musk's Grok AI chatbot.

👏👏👏👏👏

https://www.bbc.co.uk/news/articles/cq845glnvl1o

Which women are attacking ofcom?

Ofcom female employees are being attacked as a way of attacking Ofcom. Digitally stripped of their clothes by Grok, humiliated and sexually exploited on X.

They are being used.
As proxies.

As proxies for the target (Ofcom) may be, but the ultimate target is X and by attacking the female staff at Ofcom they can get Ofcom to act as a their proxy in attacking X. Two sets of proxies.

So you think X is the victim in all this?

Not the women complaining about being stripped by Grok and sexually exploited on X making complaints? Not the female employees of Ofcom who've been stripped by Grok and humiliated on X for actioning those complaints?

You think the real victim in all this is X, a business owned by the richest man in the world? I get that you're not a feminist. I really really get that.

No women are - women who are only able to discuss their oppression under gender ideology on X when even on here MN censors our speech, women in Iran where the rebellion was only mentioned on MSM after two weeks of discussion on X and news of whose deaths are now smuggled out by Starlink, women who were abused in grooming gangs and shut down on other platforms for ‘racism’ in describing their attackers, women constantly silenced on other platforms and banned or had their replies removed, Yazidi women whose description of sexual slavery at the hands of ISIS is called Islamophobia, lesbian women who are silenced elsewhere for refusing to call men ‘lesbians’, Christian women massacred in Nigeria and Sudan by Islamists whose plight is ignored by MSM, women locked up in prisons with male rapists whose campaigners can only discuss on X, women exploited by sex work or surrogacy where criticism is frowned upon on other platforms, women who have had their images stolen and manipulated on one or more of the 35,000 other AI image manipulators that are ignored by government, women who were groomed as girls on Snapchat or Meta that the government don’t care about….

So glad you got there in the end @BrokenSunflowers it's women who are the victims. Women and children. Not a business.

I think independent of platform with AI becoming better and more prevalent there maybe some from acceptance by some that their images are going to be used for pornographic purposes in life. This is really depressing.

The interest in Grok and the fact this issue is attracting parliamentary time shows how widespread this issue is. I was watching a news item on this and interviews with women suggest many are now not displaying photographs on SM and are reluctant to have their images taken generally.

is this going to lead to a generation of hidden women where women won't want to be photographed for corporate work photographs? The female head teacher of a school knowing some random schoolboy will be having a laugh producing naked (or worse) images of her and spreading the images around. Sports women would be another obvious target.

So you think we can legislate for this or is this a new depressing result of technology introduction without thought?

so you agree people are using ofcom as a proxy to attack free speech on x that will harm women and children because they want to shut them up and because they dislike musk for allowing criticism of the misogynistic far left ideologues on it.

Grok is pretty irrelevant though to this generally though, and a late comer. It is simply one more platform. Everyone I know was becoming a lot more cautious about sharing images well before Grok came out. They also didn’t use Twitter for that - they shared images on Facebook or instagram. And child groomers didn’t use Twitter either - they predominantly used Snapchat. This was an ad that came out over two years ago before Grok existed: