The rapid expansion of digital financial services is being accompanied by a growing fraud threat for businesses and financial institutions. According to a 2025 joint study by Experian and Forrester Consulting, around 62% of businesses in India reported an increase in overall fraud attacks, while 66% of organisations said their financial losses from fraud had risen year-on-year. The study surveyed 109 senior fraud decision-makers in India.
Which Digital Fraud Threat Is Growing the Fastest?
Account takeover has emerged as the most widespread and rapidly increasing fraud threat, according to the study. Around 77% of respondents reported an increase in account takeover incidents. Money muling and identity theft followed, with 71% of respondents reporting a rise in each category.
Synthetic business fraud and first-party fraud also recorded significant increases, with 39% of respondents reporting growth in both categories. Synthetic identity fraud involves combining genuine and fabricated personal information to create identities that can potentially be used to obtain financial services, credit or other benefits.
The study found that the challenge is not limited to the growing volume of fraud. Detecting and preventing increasingly sophisticated attacks is also becoming more difficult. Authorised Push Payment (APP) fraud was identified by respondents as the most challenging type of fraud to detect and prevent, followed by identity theft and money mule activity.
Why Are Fraud Attacks Becoming Harder to Detect?
First-party fraud, synthetic identities and deepfakes are adding further complexity to the fraud landscape. Criminals can use fabricated digital identities and artificially generated content to create the appearance of legitimate customers or businesses. This makes it increasingly difficult for conventional, rules-based fraud detection systems to distinguish genuine activity from malicious behaviour.
The report identified limitations in data, technology, system agility, operational processes and model performance as factors making fraud prevention more difficult. Traditional systems that depend heavily on predefined rules and established patterns can struggle when fraudsters rapidly change their methods.
The study emphasised the need for financial institutions and other organisations to shift towards more adaptive fraud prevention strategies. Instead of relying primarily on static, rules-based controls, organisations need systems capable of analysing data, behavioural signals and relationships between accounts and transactions. Such approaches can improve risk assessment during customer onboarding and help identify suspicious activity at an earlier stage.
Which Lending Products and States Show Higher Fraud Risk?
Analysis of application anomalies across lending products also showed that fraud risk varies significantly by product. Credit card applications consistently recorded the highest anomaly rates, although the rates stabilised after an initial decline. Business loans showed a gradual downward trend, with a temporary increase around the first quarter of financial year 2026.
Auto loans showed steady improvement, with anomaly rates declining across most quarters. Personal loans remained comparatively stable, with only minor fluctuations. Two-wheeler loans consistently recorded the lowest anomaly rates among the lending products covered in the analysis.
The study also identified significant geographical differences in application anomalies. Although anomalies were recorded across all states, the incidence varied considerably. Delhi, Haryana, Rajasthan, Uttar Pradesh and West Bengal recorded catch rates above 10%. Kerala, Tamil Nadu and Karnataka reported comparatively lower rates of below 8%.
Application anomalies and mule activity have emerged as major concerns because criminals can exploit weaknesses in identity verification and customer onboarding processes. Once fraudulent or compromised accounts are established, they can be used as part of networks designed to receive, move and withdraw proceeds from financial crimes.
How Are Deepfakes and AI Changing Fraud Prevention?
Cybersecurity experts said financial institutions can no longer rely on identifying individual suspicious transactions in isolation. They increasingly need to analyse customer identities, devices, transaction behaviour, account activity and links between connected accounts as part of a broader risk assessment framework.
The growing use of deepfakes and AI-generated content makes this shift even more important. Fraudsters can increasingly create convincing identities, documents and communications, making conventional verification methods less reliable. Financial institutions therefore need stronger identity controls, behavioural analytics and network-based monitoring to identify suspicious activity before significant losses occur.
The findings indicate that fraud prevention is undergoing a fundamental transformation. As digital financial activity expands, data-driven and adaptive detection systems are becoming an increasingly important layer of protection for institutions seeking to limit financial losses and maintain customer trust.
