- Interpol says GenAI-powered fraud 4.5x more profitable
- AI boosts phishing, deepfakes, and social engineering campaigns
- Agentic AI could enable autonomous end-to-end fraud in future
Cybercriminals and fraudsters using Generative Artificial Intelligence (GenAI) are 4.5 times more profitable than those not using it, Interpol is saying.
In a new research paper, titled “Global financial fraud threat assessment”, the international law enforcement agency said AI, “greatly boosts both efficiency and effectiveness” of scam campaigns, suggesting that its popularity in the criminal underbelly is only going to grow.
There are numerous ways in which crooks can use GenAI, but the most obvious one seems to be - polishing phishing content. Before the emergence of AI, the best way to spot a phishing email was to simply proofread it, since the fraudsters were usually non-native speakers, and the messages were riddled with mistakes that made it obvious they didn’t come from legitimate brands.
Abusing Intune
With AI to polish and rephrase the content, proofreading is no longer a viable option, and phishing emails became more successful and impactful.
But that’s just the “gateway drug” to AI-powered fraud. High-level crooks are using AI for deepfakes, creating hyper-convincing voice clones from almost no source material.
To make matters even worse, the dark web is full of widely available kits (deepfake-as-a-service) that further lower the barrier for entry and make kicking off an impersonation campaign just a matter of dollars.
"Over the past two years, technology has continued to enable and enhance financial fraud, empowering criminal networks to scale operations exponentially with minimal investment," said Interpol. "Digital technology and AI, in particular, have dramatically transformed social engineering techniques and victim profiling, enabling fraudsters to construct highly persuasive fraud environments.
Interpol also discussed Agentic AI - systems that “can autonomously plan and execute complete fraud campaigns, from reconnaissance, to ransom demands.” For crooks it sounds promising, but it has not yet reached the level of mass-use as GenAI. Whether or not that happens remains to be seen. After all, the promise of agentic AI has yet to fully materialize in the legal world, too.
Via The Register
