Banking sector utilizes AI to balance fraud detection and transaction speed

The banking sector faces a significant technical challenge in utilizing Big Data and Artificial Intelligence to detect fraud without disrupting legitimate consumer transactions. According to Arturo Peralta Martín-Palomino, a professor at the Universidad Internacional de Valencia, the goal is to identify ‘weak signals’ of criminal activity rather than just blocking unusual patterns.

Sophisticated cybercriminals now use fragmentation and dispersion strategies to move capital. To counter this, AI models must analyze entire networks of transactions and behavioral patterns rather than isolated events. A major risk in this engineering process is the potential for overly rigid algorithms to block legitimate purchases, such as travel or unusual but valid payments, which can cause significant economic and reputational damage to banks.