X bans Grok AI's ability to digitally undress real people
A much-needed step has been taken.
X has updated the image generation policy of its AI chatbot, Grok, which previously allowed it to produce fake images of real people in sensitive scenarios.
Elon Musk's AI tool, Grok, recently received an update to its image generation capabilities, allowing you to use its reply bot to edit uploaded images. For instance, you can ask the bot to edit an image of an iPhone 17 and make it look like a Pixel. That's a very basic use case for this new upgrade, as some notorious users on X have used the feature to produce sexualized images.
If I had to be more direct, a "put her in a bikini" prompt exploded on the platform, where users were leveraging Grok's image-editing power to strip clothes from photographs of women and show them in bikinis. The most dangerous thing about this was that it gave everyone the power to edit anyone's images and put them into fake sexual scenarios. Users even used this feature to edit images of high-profile individuals like Bill Gates, Swedish Deputy Prime Minister Ebba Busch, and Elon Musk himself, making them wear inappropriate clothes.
A lot of damage has already been done
As expected, the chatbot received a lot of backlash because of this. Both Apple and Google were recently requested to remove the X and Grok AI apps from the App Store and Play Store, respectively. A British media regulator named Ofcom reportedly began an investigation to check whether the sexually explicit deepfake images generated by xAI violate the country's Online Safety Act framework.
The IT ministry of India also sent a formal notice to X asking for the removal of all the sexualized fake images, along with a detailed report describing all the actions the company is taking to control such obscene image generation. Both Malaysia and Indonesia took the bold step of banning the Grok AI tool in their respective countries.
Many popular figures, like JerryRigEverything, criticized Elon Musk for this feature and called for cleaning the platform of such deepfake images. After all this, X has finally announced that it has implemented the necessary measures to ensure Grok AI can no longer edit images of real people in revealing clothes. In addition to this, the image creation ability, which until now was available to all users, has been made exclusive to paid subscribers.
The social media platform has also implemented geoblocking. It basically means that if your country doesn't allow it, then you won't be able to use Grok to produce AI images of real people in any kind of revealing clothes.
The internet has been flooded with AI-generated images ever since AI chatbots like Grok, Gemini, and ChatGPT gained image-generation capabilities. Give a simple prompt, and the chatbot will generate an image according to it within a few seconds. The recently popular Ghibli-style trend was part of this, where users leveraged ChatGPT's image generator to create cute Ghibli-style artwork from their photos. However, the recent Grok controversy has shed light on the dark side of this fun experience as well.
These chatbots have given everyone who knows how to write prompts the power to edit images of people in any form they want. In fact, these tools have become so advanced recently that it's very difficult to tell if an image is generated using AI. And that's not a good thing. This will make it easy to spread fake information and create deepfake images of individuals doing things that could ultimately harm their reputations.
Let's be real: AI is here to stay, and banning these tools is definitely not an option, as they have plenty of good use cases too. Instead, authorities could take necessary steps to monitor the kinds of images they generate. Similar to what happened with xAI, action should be taken against Gemini, ChatGPT, or any other AI tools if they are found to easily allow the generation of such sexual images of real people.
Follow us on Google News
Popular stories
Latest News
FCC OKs Cingular\'s purchase of AT&T Wireless
Things that are NOT allowed:
To help keep our community safe and free from spam, we apply temporary limits to newly created accounts: