Jumio’s global survey reveals deepfake awareness and identity fraud concerns related to generative AI technology. Learn how customer overestimation amplifies vulnerability and explore Jumio’s IT security solutions.
The second installment of Jumio’s yearly global consumer study has been released. Jumio is a provider of automated, end-to-end identity proofing, risk assessment, and compliance solutions. The findings from this year show that consumers are aware of how generative AI and deepfake technologies could hasten identity fraud. Although consumers recognize the need of IT security, the survey also found that they overestimate their capacity to recognize deep fakes, which might make them even more susceptible to assault.
8,055 adult consumers were studied for the Jumio 2023 Online Identity Study, which was carried out by Censuswide. The consumers were evenly distributed across the United Kingdom, United States, Singapore, and Mexico. 67% of respondents claim to be familiar with generative AI technologies like ChatGPT, DALL-E, and Lensa AI. All of which are capable of creating fake content, such as audio, video, and photos. Consumers in Singapore (87%) and the UK (56%) had the highest and lowest levels of awareness, respectively.
Underestimating how advanced the technologies are
Consumer awareness of generative AI and deepfakes is high; 52% of respondents think they could recognize a deepfake film. This attitude is a result of consumer overconfidence. Particularly considering the fact that deepfakes have become so sophisticated that they cannot be seen with the naked eye.
This is troubling given that, according to recent data from UK Finance, impersonation scams cost the UK £177 million in 2022. The study emphasized precisely how scammers have been driving this trend by becoming more difficult to detect as warning indicators. For instance, the usage of generative AI technologies has reduced the prevalence of mistakes and fake-looking websites.
Jumio data also reveals a consistent increase in the use of deepfakes across the globe and in many industries. The payment and cryptocurrency industries have a noticeable presence.
Many people appear to believe they can recognize a profound phony. Even though there are unmistakable warning indicators, deepfakes are improving tremendously and are getting harder to see without artificial intelligence (AI), according to Stuart Wells, chief technology officer of Jumio.
“While businesses will increasingly need AI-powered technology to detect deepfakes and secure their networks and clients from them, consumers may protect themselves by being wary of suggestive images, videos, and audio. Usually, a fast search will reveal whether it’s a fake or not.
Understanding of potentially dangerous use replaces awareness
There is a growing understanding of how these technologies could be exploited to facilitate identity theft as people become more aware of them. The majority of consumers (57%) think that online identity theft will grow simpler as a result, while Singaporeans had the highest awareness of their possible negative uses (73%). These percentages decline among customers in Mexico (62%), the US (49%), and the UK (43%).
Businesses must spread awareness and improve protection
“Organisations have a duty to educate their customers on the nuances of generative AI technologies to help them develop more realistic expectations of their ability to detect deepfakes,” said Philipp Pointner, Jumio’s head of digital identity. However, no amount of education, no matter how good, will ever be able to stop a fraudster from using new technologies.
“Online businesses need to put multimodal, biometric-based verification systems that can catch deepfakes into place. They must also stop the misuse of stolen personal data. Our research revealed a considerable customer interest for this type of identity verification, which is encouraging. As a result, companies should move quickly.
According to the survey, 68% of customers are open to using a digital identity to prove their identify online. Financial services (43%) and government (38%) are the top industries where people would choose a digital identification over a physical ID (such as a driver’s license or passport).