It has been a day since the new AI image generator of CHATGPT has been broadcasted, and social media feeds are already in Japanese animation studio, a Japanese animation studio preferred by a blockbuster movie like “My Neighbor” and “Spirited Away”. It has been flooded.
In the last 24 hours, we have seen the image of the Studio Ghibli version of ELON Musk, the Lord of the Rings, and the AI creation image representing President Donald Trump. Openai CEO Sam Altman seems to have even made his new profile picture with a studio Ghibli style image made of the default image generator of GPT-4O. Users seem to upload existing images and photos to CHATGPT and request to rewrite the chatbot into a new style.
Openai’s latest updates are offered on a heel of Google’s similar AI image in the Gemini Flash model, and in early March, people used to remove watermarks from the image.
With the latest tools from Openai and Google, you can enter a text prompt to rewrite the copyrighted work style. This new AI image function seems to be re -evaluating concerns about the core of various lawsuits against the generated AI model developers. If these companies are being trained for copyrighted works, are they violating copyright law?
According to Evan Brown, an intellectual property lawyer of NEAL & MCDEVITT, a product such as GPT-4O’s default image generator is operated in a legitimate gray area today. According to Brown, the style is not explicitly protected by copyright.
But Brown says that Openai has achieved this similarity by training a model for millions of frames in Ghibli’s films. In this case, many courts still decide whether to use the AI model for copyrighted works for fair use.
In an interview, Brown said, “I think this is the same question as we have been asking ourselves for several years.” “What is the effect of copyright infringement that crawls the web and copies it to these databases?”
New York Times and some publishers filed an active lawsuit against Openai, claiming that the company taught AI models for copyrighted works without proper properties or payments. Similar arguments for other major AI companies, including META and AI image creation startup, Midjourney.
A spokesman for Openai says in a statement on TechCrunch, Chatgpt refuses to replicate “the style of individual living artists”, but Openai can duplicate the “larger studio style.” Of course, it is worth noting that there is a living artist who is recognized as pioneering a unique style of studio such as Hayao Miyazaki, co -founder of Studio Ghibli.
Obviously, the user was able to recreate style from other studios and artists using the default image creation function of GPT-4O. Another person created a portrait of Marc Andressen in the style of Dr. Seuss, and the married couple made a wedding photo in a pixar style.
We tested some popular AI image generators, including things available in Google’s Gemini, XAI ‘S Grok, and Playground.Ai, and confirmed the ability to match the style of Studio Ghibli. Openai’s new image generator has created the most accurate replica of animation studio style.
Currently, the new image functions of Openai and Google suggest a leap on what AI models can create, which seems to be rapidly increasing. Openai delayed the rollout of a new image tool to users with a high demand on Wednesday. It is the most important thing for these companies today, but we must wait for the court to put weight on legality.
