Why is ChatGPT still refusing to make images of real people? I see people post them all the time

Understanding ChatGPT’s Limitations with Generating Real-Life Person Images: Policies and Practices

In recent discussions within the AI community, a recurring topic involves ChatGPT’s image generation capabilities, particularly regarding the creation of images depicting real people. Despite reports of users successfully sharing such images, many users with ChatGPT Plus still encounter strict refusals when requesting portraits or likenesses of recognizable individuals. This phenomenon raises questions about the underlying policies, the consistency of enforcement, and whether shifts in guidelines have occurred.

ChatGPT’s Image Generation Policy

OpenAI has established clear policies that restrict the use of its AI models for generating images of real people. These restrictions are primarily rooted in ethical considerations, privacy concerns, and the potential for misuse, such as the creation of deepfakes or misleading content. Consequently, when users attempt to generate images of celebrities, public figures, or private individuals, ChatGPT often declines, citing policy violations.

Reports of Policy Evolution

Some community members suggest that OpenAI’s policies have become more flexible over time. There are anecdotal claims indicating that earlier strict restrictions might have been relaxed or that specific scenarios now receive different treatment. However, OpenAI maintains a cautious stance, emphasizing user safety and ethical use, which continues to inform the core guidelines.

Discrepancies and Potential Workarounds

Despite official policies, there are instances where users see others posting images resembling real individuals. These variations may be due to several factors:

• User Workarounds: Some users employ prompt engineering techniques, indirect descriptions, or artistic styles to influence the AI’s output, sometimes circumventing restrictions.

• Community Sharing: Community platforms may share or host generated images outside the official OpenAI interface, sometimes less constrained by moderation.

• Differences in Implementation: Regional or account-based enforcement may vary, leading to inconsistencies in what content is permitted.

Conclusion

While ChatGPT and related tools aim to balance innovation with ethical responsibility, users should be aware of the existing limitations imposed to prevent misuse. As policies evolve, OpenAI continues to prioritize user safety, which may reflect in ongoing restrictions. For those seeking to generate images of real people, understanding these boundaries and respecting privacy and ethical considerations remain essential.

Disclaimer: Always adhere to the terms of service of the AI platforms and respect individual privacy rights when creating or sharing generated content.