Since Openai launched ChatGpt’s Ghibli-style AI image generator last week, it has taken social media by storm. From politicians and celebrities to everyday users, everyone appears to share an artificial intelligence-generated portrait of the legendary Miyazaki Hayao of Ghibli legend. The latest version allows people to turn their photos, or viral internet memes, into amazing Ghibli-style artwork.
Elon Musk’s AI chatbot Grok also features new Ghibli-style image generation capabilities. Elon Musk’s Xai integrates this feature with Grok 3, allowing users to create Ghibli-inspired images for free.
But not everyone is on board. Digital privacy activists on social media platform X claim they may be using the trend as a way to collect thousands of personal images for AI training. While users enjoy this feature, critics have unconsciously handed over fresh face data to Openai, warning that it could potentially raise serious privacy concerns.
Also Read | Free ChatGpt users can now create studio Ghibli style AI images. This is the way
This trend highlighted ethical concerns about artificial intelligence tools trained in copyrighted, creative works, and what it means for the future livelihoods of human artists. Miyazaki (84), known for his hand-drawn approach and whimsical storytelling, has expressed his skepticism about the role of AI in animation.
However, activists have urged Openai’s data collection strategy not just a matter of AI copyright. According to them, it allows the company to obtain voluntarily submitted images and bypasses legal restrictions applied to web-scraped data.
Also Read | How to create Ghibli-style AI images using GROK 3? Here’s a step-by-step guide
Under GDPR regulations, OpenAI must justify scraping images from the Internet under the legal basis of “legitimate interests.” This means that additional protection measures must be implemented to protect user privacy and ensure compliance. This includes demonstrating that data collection is necessary and that it does not override individual rights and follows strict measures of transparency and accountability.
In the long X-Post, AI co-founder Luiza Jarovsky of Tech & Privacy Academy said that when people voluntarily upload these images, they agree to Openai to process them (Article 6.1.A of the GDPR). This is another legal basis for giving Openai more freedom, and the legal profit balance test is no longer applicable.
Additionally, Openai’s privacy policy explicitly states that users are collecting personal data input to train AI models when users are not opting out,” she writes.
Luiza Jarovsky further stated that Openai has free and easy access to personal or private images, only they have the original. Social media platforms and other AI companies can only see the “Ghibliid” version.
Furthermore, this trend is ongoing and when people want their own fun avatars, they are learning that they can simply upload photos to ChatGpt. They no longer need a third-party provider for that,” she said.
The Himachal Cyber Warriors, who claimed to be a team of cybersecurity experts, said, “Think before #ghibli. Think about that cute Ghibli-style selfie. Your photos can be misused or manipulated. AI may train them without your consent.
What ChatGpt said
Openai has not yet issued a statement regarding the safety of your personal data using the Ghibli-style AI Image Art version. However, when HT enters a prompt to ask ChatGPT if it is possible to safely upload personal photos to Ghibli Art Generator, he says, “No, it’s not safe to upload personal photos to AI tools unless you’re sure about privacy policy and data processing practices. An app specifically designed for secure image processing.”
What Glock said
Q. Is it safe to upload personal photos to Grok 3 for Ghibli style AI images?
Grok’s reply to the prompt: “I haven’t explicitly explained how long Xai will keep uploaded images or if it will be used to train future models, but the wider AI landscape provides clues. X’s post has no hard confirmation from Xai as of March 31, 2025, but X’s photos can be used by default. Compare this to CHATGPT.
“There’s a corner of misuse. It’s theoretically possible for AI to replicate your portrait in other contexts once the photo is processed, but Grok 3 does this intentionally. Data leaks are another risk. Passing fresh facial data over to the AI system.”
“On the other hand, if you’re just uploading casual, low-stakes photos, such as pets or landscapes, and you’re likely to leave it somewhere on the server, the risk is minimal. Roll the dice,” he added.
How to Keep Your Biometric Data Safe
Consider the risks before jumping to the next AI image trend. Here’s how you can protect your privacy:
Think carefully before uploading personal photos of images generated by AI. Avoid sharing high-resolution images on social media. It can be trimmed for AI training. You can use a pin or password instead of facial recognition to unlock your device.
