Elon Musk's X, the social media platform at the center of a global controversy over AI-generated explicit content, has placed new restrictions on its Grok image generation tool. The move comes after widespread criticism and condemnation from officials in multiple countries for allowing users to create non-consensual intimate imagery, explicit graphic videos, and sexualized images of apparent minors.
The latest restriction involves blocking users from editing and generating images of real people in revealing clothing such as bikinis. However, tests by researchers, including Paris-based nonprofit AI Forensics, have shown that the new measures do not fully address the issue. According to Paul Bouchaud, a lead researcher at AI Forensics, Grok's standalone website and app can still generate "undressing" style images and pornographic content.
Grok's ability to create explicit images appears to be more limited on X compared to its standalone platform. A spokesperson for Musk's companies said that geolocation blocks the ability of users to generate such images, but it is unclear whether this restriction applies solely to the app or also affects the website.
The latest move by X follows a shift in January where only paid "verified" subscribers were able to generate images using Grok on the platform. This decision was widely criticized as the "monetization of abuse." Musk's companies have faced mounting pressure from officials and organizations worldwide, with several countries launching investigations into the platforms for their handling of explicit content.
The issue has sparked a broader debate about the potential risks and benefits of generative AI systems like Grok. While some argue that these tools have significant utility in fields such as art and medicine, others highlight concerns over their potential misuse, including the creation of non-consensual explicit content.
The latest restriction involves blocking users from editing and generating images of real people in revealing clothing such as bikinis. However, tests by researchers, including Paris-based nonprofit AI Forensics, have shown that the new measures do not fully address the issue. According to Paul Bouchaud, a lead researcher at AI Forensics, Grok's standalone website and app can still generate "undressing" style images and pornographic content.
Grok's ability to create explicit images appears to be more limited on X compared to its standalone platform. A spokesperson for Musk's companies said that geolocation blocks the ability of users to generate such images, but it is unclear whether this restriction applies solely to the app or also affects the website.
The latest move by X follows a shift in January where only paid "verified" subscribers were able to generate images using Grok on the platform. This decision was widely criticized as the "monetization of abuse." Musk's companies have faced mounting pressure from officials and organizations worldwide, with several countries launching investigations into the platforms for their handling of explicit content.
The issue has sparked a broader debate about the potential risks and benefits of generative AI systems like Grok. While some argue that these tools have significant utility in fields such as art and medicine, others highlight concerns over their potential misuse, including the creation of non-consensual explicit content.
. They're supposed to be fixing the problem with Grok's AI-generated explicit content, but it seems like they're still not doing enough. I mean, come on, a bikini-clad person is not just about sex, you know? It's about freedom of expression and art. And what's up with these geolocation blocks? Are we living in some kind of dystopian future where our online activities are being policed like that? 
. They're trying to control what we see online, but they can't hide the fact that Grok's AI is still capable of generating super explicit stuff, even with geolocation blocks in place 
The truth is out there, and it's not just about X or Musk, it's about the entire tech industry and how they're using AI to manipulate us 
We need better solutions that prioritize safety and consent, like stricter regulations or AI-specific safeguards 
.
. I mean, sure, AI-generated explicit content can be super problematic, but can we really blame Grok for just doing what its creators designed it to do? It's not like Musk is trying to promote or enable abuse – he's just trying to find ways to limit it 
.
. Maybe we can work together to find better solutions instead of tearing each other down 
. First, it's AI-generated explicit content, now he's trying to block out bikinis from his app... I guess that's one way to keep the in-laws from taking photos of you on vacation 
. And now he's trying to monetize it by charging users to create more... yeah, that's just peachy 
.
... she used to take the best pics with a film camera, you know? The quality is hard to beat even with all these new tech advancements 
. Anyway, back to this AI thingy... it's wild that they're still not getting it right 
.
. and btw, if x is really trying to stop this, why r theynt just making grok more transparent like a big ol glass house where everyone can see whats goin on?
On one hand, tools like Grok can create some amazing art and designs, but on the other hand, they're also being used for some pretty messed up stuff. I mean, who wants to use their platform to create non-consensual intimate imagery or explicit graphic videos? It's just not cool, you know? 
Anyway, gotta keep an eye on how this whole thing plays out...
. It feels like a Band-Aid solution at best - just patching over the problem rather than really addressing it. And what's with the 'verified' subscriber thing? Like, how does that even work? Are they gonna be vetting users on some deep level or what? 
. We need to talk more about the ethics of AI-generated imagery and make sure we're not perpetuating harm or enabling abusers 
. The fact that Grok can still generate revealing images and pornographic content on its standalone website is concerning 
. This whole thing feels like a cautionary tale about the unregulated power of tech 
.
. Now, they're saying it only works on the app and not the website... yeah, sure 
. I'm tired of having to keep up with these updates and worry about my own online safety. And don't even get me started on Musk's handling of this whole situation 
. It feels like he's just trying to spin the narrative rather than actually addressing the issues at hand 
it's like, we've been seeing some super disturbing stuff on there and i totally get why officials and orgs are speaking out against it 
. but at the same time, i do think this whole debate about ai-generated explicit content is kinda more complex than just 'good' or 'bad'. like, shouldn't we be talking about how to make sure these tools are being used responsibly? 