Be Alert AI and Ghibli Users! ChatGPT Can Generate Your Fake ID Cards including Aadhar Card and Pan Card!

Last week, OpenAI released the native picture generating features of GPT-4o, or more specifically, ChatGPT. Over 700 million photographs have been produced by users until Friday, discovering a variety of applications for the new AI tool, most notably the production of portraits in the style of Studio Ghibli. However, the possibility of misuse increases as more people become aware of ChatGPT's new features.

Along with a correct QR code and Aadhaar number, other users on X (Twitter) also posted pictures of Tesla Chief Executive Officer (CEO) Elon Musk and OpenAI CEO Sam Altman superimposed on the Indian ID, Aadhar card.

Using ChatGPT's new image generator, some social media users have begun posting pictures of fake Aadhaar cards with their own photos. With ChatGPT's recent feature to generate real graphics, the long-running discussion over AI companies releasing features that could be misused in the wrong hands.

We attempted to use ChatGPT to create an image that resembled an Aadhaar card after seeing the deluge of photographs on social media. Except for the facial characteristics, the results were extremely similar to the actual Aadhaar card image.

Not Only Aadhaar Cards: ChatGPT Can Also Generate Fake PAN Cards

Misuse of ChatGPT is not restricted to Aadhaar cards; social media users have reported cases where the chatbot can accurately create PAN cards as well.

The Unique Identification Authority of India (UIDAI) issues Aadhaar cards, whilst the Income Tax department issues PAN cards.

Are There Any Restrictions for Image Generation in ChatGPT?

Erotic content, violent, abusive, and hostile imagery, as well as photorealistic photographs of children (even minor public figures) are currently strictly prohibited on ChatGPT.

ChatGPT's native image generating capabilities imply that the underlying foundation model that powers the chatbot can generate images directly, rather than relying on external models such as DALL-E 3. This enables the chatbot to follow specific natural language commands, resulting in more complex and accurate visuals.

OpenAI acknowledged in their GPT-4o Native Image Generation system card that the new model may pose greater hazards than earlier DALL-E versions.

The ChatGPT's parent company OpenAI stated, "Unlike DALL·E, which operates as a diffusion model, 4o image generation is an autoregressive model natively embedded within ChatGPT. This fundamental difference introduces several new capabilities that are distinct from previous generative models, and that pose new risks... These capabilities, alone and in new combinations, have the potential to create risks across a number of areas, in ways that previous models could not."