Artificial intelligence (AI) chargeback fraud: An emerging threat

Despite technological leaps in fraud detection and prevention tools, fraud is still wreaking havoc on businesses globally. One survey found that 43% of respondents expect fraud activity to rise in 2024. Some have hailed AI as a saviour in this space. Unfortunately, AI is also available to fraudsters.

The AI fraud threat

Like anything, AI has its positives and negatives. However, a new AI-enabled threat looms, threatening travel providers’ bottom lines. Now, chargeback fraud is nothing new. What is new, however, is fraudsters using AI to create automated false chargebacks. This sophisticated method allows fraudsters to operate at a considerably larger scale to create false chargeback claims that are also less likely to be detected. It’s far from an ideal situation for busy travel providers already facing a high rate of fraud.

How large language models (LLMs) are enabling large-scale fraud

LLMs, like ChatGPT and Gemini, allow fraudsters to commit chargeback fraud. However, it’s not the quality of the output but the quantity that allows them to make money from chargeback fraud. This is because LLMs enable fraudsters to significantly increase the number of false chargebacks they raise. 

Here’s how it generally works: Imagine a fraudster using ChatGPT to generate hundreds of slightly different false identities, each with a different backstory. These false identities are used to create bookings and file chargebacks with false complaints. ChatGPT and other LLMs can also create convincing responses to card providers' questions, making the chargeback claims seem legitimate on the surface.

In short, LLMs allow fraudsters to create identities using the stolen information required to make chargeback claims at scale. They can file several hundred chargeback claims daily, meaning that even if many of the claims are rejected, the fraudsters still make a lot of money at the expense of banks and travel providers.

As previously mentioned, organised chargeback fraud isn’t new. However, pre-LLMs, this method was hard to do at scale due to the admin involved. After chargebacks are filed, fraudsters may be required to respond to requests from card providers—and provide highly accurate information or risk being found out. LLMs allow fraudsters to automate aspects of this process. 

To make matters worse, how fraudsters use LLMs will get more sophisticated over time, making sufficient chargeback fraud protection necessary for travel businesses of all sizes.

Tackling AI chargeback fraud

A considerable amount of time and money is funnelled into making sure companies and fraudsters are in lockstep—on both sides. Fraudsters adopt advanced technology, like AI, to extract as much money as possible. 

Companies, on the other hand, pour resources into developing powerful fraud protection tools to detect and prevent increasingly sophisticated fraud attacks. It’s a continuous cycle of adaption and innovation on both sides.

Due to increasing fraud, including AI chargeback fraud, more travel businesses are investing in chargeback support services. The good news is that you don’t necessarily have to engage with a service that only offers chargeback support. Companies like Trust My Travel combine several vital services into one handy subscription.

Chargeback support is just the beginning

Trust My Travel subscribers receive chargeback support from our experienced fraud specialists. But that’s not all—our pay-as-you process travel payment solution combines a merchant account, financial failure insurance, supplier payments and chargeback support. The best bit? There’s no bond required! 

Are you ready to start protecting travel payments? Get in touch with us today to find out how we can help your travel business thrive.