Sunday, 24 August 2025
28.8 C
Singapore
28 C
Thailand
19.9 C
Indonesia
27.9 C
Philippines

OpenAI pauses free GPT-4o image generation after viral Studio Ghibli trend

OpenAI halts free GPT-4o image generation after viral Studio Ghibli trend raises legal concerns, leaving paid users with continued access.

OpenAI has temporarily suspended free access to its in-app image generator, powered by the GPT-4o reasoning model, just a day after its release. The decision follows a viral trend where users flooded social media with AI-generated images mimicking the style of Studio Ghibli, the renowned Japanese animation studio.

The update, designed to enhance realism in both text and images, quickly gained traction as users experimented with the technology. Many shared AI-generated pictures reimagining personal family photos in Studio Ghibli’s signature style. Others recreated famous scenes from the 2024 Paris Olympics, classic films such as The Godfather and Star Wars, and internet memes like the “distracted boyfriend” and “disaster girl.”

OpenAI CEO reacts to the viral trend

Even OpenAI CEO Sam Altman joined the trend, updating his X profile picture to a Studio Ghibli-style version of himself. However, he soon announced on the platform that OpenAI would delay the image generator update for users on the free tier of ChatGPT. He did not specify whether or when the feature would return. The update remains available to paid subscribers on the Plus, Pro, and Team plans.

GPT-4o’s image generation uses an “autoregressive approach,” meaning images are created gradually from left to right and top to bottom instead of appearing all at once. This process helps improve accuracy and realism, particularly for AI-generated text, which is often blurry or unreadable in older models. OpenAI also told The Wall Street Journal that it had fine-tuned the model through human-led post-launch training to reduce standard text and image rendering errors.

The use of AI to replicate artistic styles has long raised legal and ethical questions. OpenAI previously disclosed the sources used to train its AI, but users’ interactions with the model remain unpredictable.

Intellectual property lawyer Evan Brown said that generating Studio Ghibli-style images does not necessarily break copyright laws, as “style is not explicitly protected by copyright.” However, it exists in a legal grey area. OpenAI has clarified that while ChatGPT can replicate a studio’s aesthetic, it does not emulate individual artists’ style.

To avoid potential legal issues, OpenAI has removed GPT-4o’s image generation from the free ChatGPT tier while considering its next steps. This is not the first time AI-generated content has sparked controversy. A previous AI trend in late 2023 allowed users to create Funko Pop! figures in Microsoft Designer. While many generated figures of themselves, others used the tool to recreate well-known figures such as Elon Musk and Beyoncé. Users had to use creative nicknames like “Elon X” and “Queen Bey” to bypass restrictions.

As AI-generated art continues to evolve, companies like OpenAI must navigate the fine line between innovation and copyright concerns. Whether or not GPT-4o’s free image generation will return remains uncertain, but for now, only paying users can access the tool.

Hot this week

Google moves closer to nuclear power deal with Kairos and TVA

Google partners with TVA and Kairos Power on a new reactor in Tennessee, aiming to supply data centres with nuclear energy by 2030.

AfricAI joint venture launched to advance sovereign AI in Africa

AfricAI, a new joint venture, aims to advance Africa’s digital sovereignty with AI solutions tailored to regional needs.

China hosts the first world humanoid robot games with record-breaking performances

China hosts its first World Humanoid Robot Games, where 280 teams competed in events ranging from running to cleaning, achieving record-breaking results.

White House launches official TikTok account despite ongoing ban threat

The White House has launched its first TikTok account despite the looming 17 September deadline that could see the app banned in the US.

iPhone 17E expected to bring new design, Dynamic Island and A19 chip

Apple’s iPhone 17E could launch in early 2026 with a new design, Dynamic Island and the powerful A19 chip.

Google adds a shortcut to make video editing in Drive easier

Google adds a Drive shortcut for Workspace users to open and edit videos in Google Vids more easily, rolling out from 21 August.

Belkin introduces first Qi2.2 chargers with 25W wireless charging speeds

Belkin launches its first Qi2.2-certified chargers, offering 25W wireless charging speeds with three models designed for both home and travel use.

New feature begins rolling out to Windows Insiders

Microsoft is testing a Windows 11 feature to resume Android apps on PCs, starting with Spotify, for seamless cross-device use.

Meta introduces new AI-powered ad tools for holiday campaigns

Meta introduces new AI-powered ad tools to boost holiday sales, including enhanced creator partnerships, video ads and global expansion.

Related Articles

Popular Categories