by MTHULISI SIBANDA
JOHANNESBURG, (CAJ News) – SUMSUB, the verification platform, has unveiled a new adaptive deepfake detector, which the company says is designed to identify evolving fraud patterns in real time through instant online self-learning updates.
The Cyprus-headquartered firm noted that while the solution is launching globally, its relevance is especially clear across Africa, where fraudsters are shifting from low-effort scams to more sophisticated artificial intelligence-enabled attacks.
The new model pledges to tackle the prevailing issue of traditional offline solutions being unable to detect the latest deepfake scams.
Sumsub hails its deepfake detector as effectively spotting emerging types of sophisticated fraud through its machine learning (ML)-driven detection tool with instant online self-learning upgrades.
This shift is reflected in Sumsub’s Identity Fraud Report (2025–2026), which established that Tanzania recorded the highest fraud rate on the continent in 2025 at 5 percent, while Uganda recorded a fraud rate of 4,7 percent.
Ivory Coast also saw fraud rise by 51 percent year-on-year to 4,5 percent.
In Kenya, despite an overall decline in fraud, deepfakes already account for nearly 10 percent of fraud attempts, highlighting how AI-enabled fraud is becoming more prominent even in markets where traditional fraud is declining.
In South Africa, the country’s overall fraud rate declined by 31 percent year-on-year to 1,4 percent in 2025. However, deepfake incidents increased by more than 269 percent YoY, indicating that AI-enabled impersonation is quickly emerging as the next frontier in South Africa’s digital identity landscape.
“In 2026, the threat landscape has evolved, demanding risk management teams respond with next-generation fraud prevention models,” Nikita Marshalkin, Head of Machine Learning at Sumsub, said.
The official noted that modern deepfakes could no longer be detected by the human eye, and decision-making should be based on multiple signal analyses in real time.
Marshalkin said that was the reason they launched the upgraded deepfake detector, offering clients a tool and an online learning system that combined advanced document checks, device intelligence and fraudulent network analysis to complement deepfake detection capabilities.
“When the price of failure is too high, a comprehensive approach to the increasing AI-driven fraud challenge is the answer we need.”
– CAJ News