Hayao Miyazaki’s Stance on AI vs. OpenAI’s Ghibli-Inspired Image Generator

Legendary animator Hayao Miyazaki, co-founder of Studio Ghibli, has long been critical of generative AI, famously condemning it as “an insult to life itself” back in 2016. Nearly a decade later, OpenAI has introduced a ChatGPT image generator that allows users to create Ghibli-style illustrations—despite Miyazaki’s well-documented disapproval.

A Popular but Controversial Feature

The tool has seen overwhelming demand, with OpenAI CEO Sam Altman joking on X that their GPUs are overheating due to its popularity. However, as the saying goes, “If something is free, you are the product.” By default, OpenAI collects user inputs—both text and images—to train and refine its AI models.

How OpenAI Uses Your Data

Like many AI companies, OpenAI scrapes publicly available content from across the web, including articles, images, and videos, to build and enhance its models. According to OpenAI, training on a wide range of artistic styles allows its AI to generate convincing and adaptable visuals.

Yet, this practice has sparked legal battles. Many publishers, authors, and artists argue that AI companies should compensate creators whose work is used for training data. Some have filed lawsuits demanding payment for their intellectual property. If you’d rather not contribute your data to OpenAI for free, there are ways to limit its reach.

How to Stop ChatGPT From Using Your Data

OpenAI provides some privacy settings that allow users to opt out of AI training, but these options are not enabled by default. Here’s how to adjust them:

For ChatGPT:

Click on your profile picture in ChatGPT.
Go to Settings > Data controls.
Find “Improve the model for everyone” and turn it off to prevent OpenAI from using your inputs for training.
If you use the voice search feature, make sure to disable settings related to voice and video recordings as well.

For Sora (OpenAI’s Video Generator):

Click on your profile and open Settings.
Locate “Improve the model for everyone” and disable it.
To keep your content private, also turn off “Publish to Explore”, though this won’t remove anything already made public.

While these steps help limit data collection, they don’t completely prevent OpenAI from analyzing user interactions. As AI development continues to raise ethical and privacy concerns, users must decide how much personal data they’re comfortable sharing with these powerful models.