What is an AI-generated ID?
Unlike traditional counterfeit IDs, AI-generated IDs are image files that replicate the look of legitimate government-issued IDs. Because these are digital images instead of physical documents they are easier to distribute and less detectable through traditional verification methods. These fake IDs are frequently sold on the dark web for as little as $5, making them highly accessible to criminals.
AI tools can generate both front and back images of IDs, including a photo. The photo may be one provided by the user, or it could come from services like thispersondoesnotexist, which creates completely new, fake faces. Additionally, these fake IDs often have realistic-looking backgrounds to make it look like the ID is sitting on a table, countertop, or blanket.
One of the more concerning aspects of these AI-generated IDs is the service model behind them. Platforms like ProKYC provide full-service options for creating fake IDs, complete with images, videos, and the ability to bypass verification systems that use phone cameras or webcams. For less than $700 a year, cybercriminals can access these services, enabling them to generate fraudulent passports and even selfie videos for identity verification processes. These services have opened up a new avenue for ID fraud, making it more accessible than ever before.
What are AI-generated fake IDs used for?
AI-generated fake IDs are primarily used for synthetic identity fraud. Fraudsters can manipulate these IDs to commit a wide variety of illegal activities, from setting up financial accounts to engaging in scams. Some of the most common uses include:
- Account creation or access to credit at financial institutions
- Scheduling a high risk transaction in person (vehicle test drive, apartment tour, car/equipment rental)
- Account verification after lockout
- Background checks
Our findings: AI-generated fake IDs can be caught
We purchased 200 AI-generated fake ID images and ran them through our digital ID verification software.
We caught 84% of the AI-generated fake IDs with image and symbology analysis alone. When combined with third-party database checks, we were able to catch 99.6% of the fraudulent IDs.
Of the third-party database checks run, DMV checks were able to catch 100% of the fake IDs.
The importance of document tampering detection
While the quality of AI-generated fake IDs is generally high, there are still telltale signs that can expose them. In 24.2% of cases, we found evidence of photo tampering. This suggests that relying on barcode scanning or optical character recognition (OCR) alone is insufficient for detecting fraud. Document tampering detection plays a crucial role in identifying altered photos or manipulated text on an ID, something that traditional verification systems may miss.
For example, digital ID verification software might pick up inconsistencies in image lighting, shadow placement, or pixelation that would otherwise go unnoticed by humans. AI-generated fakes often leave subtle traces, like unnatural facial expressions or unrealistic background elements, which can help trained systems flag these documents.
Regional variability
The quality of AI-generated fake IDs varies depending on the state in question. For example, New York, Texas, and Arizona are frequently targeted by counterfeiters, according to our 2023 and 2024 Annual Fake ID Reports. Fortunately, these states’ IDs are among the easiest for detection software to flag—our system caught 100% of fake IDs from these high-volume states.
However, businesses must remain vigilant, as fraudsters are constantly refining their techniques to bypass current detection methods. The focus should be on a multi-layered verification approach that combines AI-powered document verification with third party database checks.
Robust digital ID verification to catch AI-generated fake IDs
For businesses verifying identities digitally, it’s crucial to have a robust system capable of catching these sophisticated AI-generated fakes. ID verification solutions must incorporate advanced algorithms, real-time data analysis, and multi-step verification processes to ensure that identities are legitimate.
Document verification solutions should check for inconsistencies in images and text, while also evaluating data patterns that may suggest tampering or forgery. Additionally, integrating third-party checks, such as DMV database checks, can significantly increase the chances of catching fake IDs.
