OpenAI has rolled out a major upgrade to ChatGPT, introducing a powerful AI image generator that has quickly gone viral for its ability to create Studio Ghibli-style illustrations. But beyond its aesthetic appeal, the real controversy lies in OpenAI’s latest shift in content moderation policies—allowing ChatGPT to generate images of public figures, racial features, and even controversial symbols under certain conditions.
The company’s latest model, GPT-4o, enhances image editing, text rendering, and spatial awareness, making ChatGPT’s visual capabilities more advanced than ever. However, OpenAI’s policy changes, which previously blocked sensitive content, now permit images of political figures like Donald Trump and Elon Musk, as well as symbols historically associated with hate—provided they are used in educational or neutral contexts.
OpenAI Shifts From Blanket Censorship to “Precision” Moderation
OpenAI’s model behavior lead, Joanne Jang, explained the reasoning behind this shift in a blog post, stating:
“We’re shifting from blanket refusals in sensitive areas to a more precise approach focused on preventing real-world harm.”
This change aligns with OpenAI’s broader plan to “uncensor” ChatGPT, giving users more freedom to generate diverse content while reducing the number of topics the AI outright refuses to handle.
Previously, ChatGPT would decline requests related to altering racial features or modifying images in ways that could be seen as controversial—such as making a character “look more Asian” or changing body types. However, OpenAI has now relaxed these restrictions, as testing by TechCrunch has shown that the AI now fulfills such requests.
Public Figures, Political Sensitivities, and AI-Generated Controversy
One of the most talked-about changes is ChatGPT’s ability to generate images of public figures, a move that OpenAI had previously avoided. The company now allows users to request AI-generated depictions of celebrities, politicians, and other well-known personalities, though it has introduced an opt-out feature for individuals who do not wish to be portrayed.
Additionally, OpenAI’s white paper confirms that the ChatGPT AI image generator can now create controversial symbols, such as swastikas, in historical or educational contexts, as long as the images do not promote extremist ideologies.
This decision is likely to fuel ongoing debates about AI ethics and content moderation—especially considering the backlash Google faced over Gemini’s AI image generator, which produced historically inaccurate racially diverse depictions of figures like the U.S. Founding Fathers.
Is OpenAI’s Policy Shift Politically Motivated?
The timing of OpenAI’s policy shift has raised eyebrows, particularly among lawmakers scrutinizing Big Tech’s role in content moderation. Republican Congressman Jim Jordan recently questioned OpenAI and other Silicon Valley firms about potential collusion with the Biden administration to censor AI-generated content.
In response, OpenAI has denied any political motivations, stating that the changes reflect a “long-held belief in giving users more control.” The company also argues that its technology has improved to the point where it can now navigate sensitive issues with greater nuance.
However, some analysts speculate that OpenAI’s loosening of content restrictions is a strategic move ahead of potential regulatory scrutiny under a future Trump administration. Other major tech firms, including Meta and X (formerly Twitter), have also relaxed moderation policies, signaling a broader industry trend toward less restrictive AI content policies.
Balancing AI Creativity and Responsible Moderation
Despite OpenAI’s relaxation of certain content policies, its new AI image generator still enforces strict guardrails in some areas. For example, GPT-4o includes stronger protections against generating images of children, a step up from DALL-E 3, the AI’s previous image generator.
However, as OpenAI pushes the boundaries of AI-generated content, the broader implications remain uncertain. Will these policy changes lead to greater freedom of expression, or will they spark new controversies over misinformation, bias, and AI ethics?
For now, the OpenAI ChatGPT AI image generator is making headlines for its viral Studio Ghibli-style art, but the real debate over AI-generated content is just beginning.
Get the Latest AI News on AI Content Minds Blog