By Nicola Mawson
AI-ASSISTED cyber-attacks, called deep fakes, are an increasing threat to global businesses. South Africa is not immune, with these techniques increasing locally.
Through the use of artificial intelligence, criminals create realistic videos, emails, and email attachments among others, to lure unsuspecting end users into clicking on the items, and gain access to a company’s network. And ICT professionals are working hard to tackle this threat, often using AI as well to detect weaknesses in IT systems so they can be plugged in advance.
While there are no definitive statistics available for such hacks in South Africa, cryptocurrency company Luno has proactively detected more than 2 000 fake social media sites with a take-down success rate of 98% and almost 500 fake websites, with a take-down success rate of 87.6% over the past year.
Globally, there was a tenfold increase in the number of deep fakes detected across every sector, according to eftsure, a payment security company.
“This dramatic rise in deep fake detection underscores the rapid advancement of AI-powered fraud techniques and the urgent need for more sophisticated detection methods,” it said.
Moreover, said eftsure, crypto is the main target sector for deep fake fraud, accounting for 88% of all deep fake cases detected in 2023. “This statistic highlights the particular vulnerability of the cryptocurrency industry to advanced fraud techniques, likely due to its digital nature and potentially high financial stakes,” it said.
A Luno staff member was recently targeted using a deep fake audio message on WhatsApp. Purportedly from a Luno executive, the instruction in the voice note raised flags and alert staff were able to avert a problem.
Deep fakes pose a material risk in the crypto sector, said Johan Hetzel, the global head of compliance and anti-financial crime at Luno, a licensed financial services provider and South Africa’s largest crypto investment app.
Luno has a dedicated team focused on keeping customers safe. It employs various tools to scan domains and social media to find instances where Luno’s brand is used in fraudulent websites or social media pages. The approach focuses on removing these platforms fast.
Francois Smit, Belgium Campus iTversity campus manager and cluster head of networking, said: “Deep fakes present a growing challenge, with advances in technology making it increasingly difficult to distinguish between genuine and manipulated images and voices, which could impact the man on the street.”
He said this had serious implications for security measures like voice recognition.
Smit said some of the attacks did, however, have characteristics that gave away the fact that they were created with AI, such as unusual speech patterns in audio files, and mismatched lip-synching, irregular shadows, and strange lighting within pictures when it comes to videos.
Arthur Goldstuck, the managing director of World Wide Worx, noted that deep fakes have been factors of misinformation, propaganda and manipulation on social media in South Africa for some years.
“The difference now is that the tools to create such content are available to all. As the tools become better, we'll also see a dramatic increase in the amount of deep fake content being distributed or used to blackmail, coerce, convince and manipulate,” Goldstuck said.
He said: “It has never been more important for individuals to be aware that attempts will be made to manipulate them. Sadly, individuals have never been more willing to be manipulated, if the messaging aligns with their world views.”
BUSINESS REPORT