Artificial intelligence (AI)-driven fraud is on the rise across Africa, with fraudsters exploiting human behaviour and trust rather than system weaknesses, says Thalia Pillay, CEO of Orca Fraud.
Speaking in an interview with ITWeb TV, Pillay says criminals are using increasingly sophisticated AI-backed tactics to bypass security controls.
The Cape Town-based financial technology company monitors more than $5 billion in monthly transaction volumes, working with major banks, telcos and payment providers across 70 markets.
“I think what people often forget is that fraud is a full-time job for fraudsters. I always say they operate like start-ups because they are very agile and innovative. They are growing very fast, and unfortunately, in this age of AI, they also have the same tools,” says Pillay.
“The big thing we are seeing is the rise of very convincing phishing scams. Where before you would be able to see punctuation errors or translation errors, these days it is hyper-realistic.”
Orca Fraud provides an AI-powered, real-time fraud orchestration and transaction monitoring platform for banks, telcos and payment providers.
Founded in 2024, the start-up raised $2.35 million (R40 million) in a seed round in March, to advance its transaction monitoring and fraud intelligence capabilities across Africa and other emerging markets.
Pillay says fraud in emerging markets is highly-localised and evolving rapidly.
“It is remarkable how different fraud looks per market. There is so much localisation because socio-economic climates are so different. In Uganda, behaviour can be very different from francophone markets or Nigeria.”
Pillay urges financial institutions and technology leaders to strengthen fraud prevention without compromising customer experience.
Highlighting common security oversights by financial institutions, Pillay says: “A lot of them focus on the perimeter. It is like putting a trailer door on the front and leaving all the windows open throughout the house.”
She adds that while many banks focus on frictionless customer onboarding and know-your-customer processes, they lack automated transaction monitoring to assess risk continuously across the entire user journey.

