Sumsub develops deepfake detector

Verification platform provider Sumsub has launched a set of models designed to enable the detection of deepfakes and synthetic fraud in visual assets.

The For Fake’s Sake suite has been made publicly accessible for download and usage. It aims to provide the AI community with a platform for testing and experimentation, while exploring ways to tackle the escalating threat of deepfakes.

The machine-learning driven models are now able to estimate how likely it is that an upload image was created artificially. Sumsub said its internal testing indicates that the AI models can accurately detect typical image alterations, and help the community to recognise AI-generated images more confidently.

According to the company’s internal data, AI-generated identity fraud cases in Australia surged 1300% in the first half of 2023 due to the rapid growth in the use of AI.

Sumsub co-founder and CTO Vyacheslav Zholudev said the company has dedicated substantial resources to combating the growing menace of deepfakes, including by developing the newly enhanced deepfake detector.

“Offering top-quality products to customers is one thing. But it is equally vital we commit to maintaining a transparent public discourse on the multifaceted threats posed by deepfakes, from impersonation of real individuals online, theft of personal data, the dissemination of misinformation, social engineering and more,” he said.

“Our open-source tool represents just one of the many strides Sumsub is taking, apart from product development, towards fulfilling our mission of empowering a safe and inclusive digital future — for companies and the wider public’s free use alike.”

Sumsub plans to continually develop the models as AI technologies advance to ensure fraudsters are not able to overcome any introduced regulations governing the use of AI technologies, he said.

Image credit: iStock.com/FotografieLink