Banks may be required to cut the time to detect and stop fraudulent transactions to as little as 30 seconds to one minute, as fraudsters increasingly move money through mule accounts at a lightning speed, industry executives said at the Global Fintech Fest 2026.
The traditional “golden window” of around 15 minutes available to banks to identify a fraudulent transaction and attempt recovery may be too long in the current environment where instant payments dominate the landscape, HSBC’s Ranjan Bhattacharya said. “30 seconds to 1 minute is the more realistic time frame,” Bhattacharya, MD, head of group strategy, Middle East and India, and chief of staff, India, HSBC, said at a panel discussion on hunting mule accounts in real time.
ALSO READ
“Today, APIs enable us to ping and take those decisions at a lightning speed. And fraudsters also are moving money very fast,” he said, adding that even two minutes could be a more realistic outer limit. The increasing use of instant digital payments such as UPI is making the window progressively shorter.
Unlike card payments, where existing dispute and transaction-monitoring mechanisms can provide some scope to intervene, UPI transactions result in an immediate movement of funds between accounts. This makes real-time detection increasingly critical. The challenge is compounded by the growing sophistication of mule-account networks. Across different forms of digital fraud—from investment scams to impersonation and “digital arrest” frauds—the money received from a victim is typically moved through multiple mule accounts before reaching the wider fraud network.
STORIES YOU MAY LIKE
India recorded losses of around Rs 22,000 crore from cyber frauds in 2025, according to government data, underlining the scale of the problem. Banks are, therefore, looking to move beyond rule-based systems towards behavioural and real-time transaction monitoring.
Devraj Sanyal, chief of AI and data science, Reserve Bank Innovation Hub, said institutions should focus less on identifying specific modus operandi and more on behavioural patterns. Once fraudsters realise that a particular method has been detected, they can simply change their approach. “If you go behind the modus, it is going to change tomorrow. What you have to go for is behavioural patterns.”
ALSO READ
The problem begins even before a transaction takes place, with mule accounts increasingly being opened through sophisticated methods. Surendra Kumar Vishwakarma, general manager at Punjab National Bank, said traditional fraud controls were increasingly being supplemented by systems that analyse transaction behaviour as it happens.
“The time has come for us to understand the behaviour of transactions,” he said, adding that banks need to act while a transaction is taking place, rather than rely only on predetermined rules. However, faster intervention also brings the risk of legitimate transactions being blocked. Banks have to balance fraud prevention with customer trust, Vishwakarma said. Panelists also pointed to the use of deepfake technology to potentially bypass video-KYC processes, making it harder for human agents to distinguish genuine customers from impersonators.
Truecaller’s Global Head — Business Shiladitya Mukhopadhyaya said network-level signals could help banks identify suspicious changes in a mobile number’s behaviour, including sudden SIM changes or a number being registered across multiple banks or UPI platforms. With fraudsters constantly changing tactics, banks may increasingly have to rely on collaboration and shared intelligence across the financial ecosystem, rather than applying fraud controls in isolation.
Read Next
This article was first uploaded on September nine, twenty twenty-six, at twenty-nine minutes past eight in the night.
© The Indian Express (P) Ltd
aniltikotekar4sme.com 