A new report shows that phishing attacks designed to steal banking credentials accounted for 23% of all employee clicks in 2025, making banking the most heavily targeted industry, ahead of the government sector at 21%.
Although employee susceptibility to phishing fell 27% year on year, an average of 87 out of every 10,000 workers still clicked on a malicious link each month, according to Netskope Threat Labs’ sixth annual threat report released on Tuesday.
As more systems migrate to the cloud, attackers are increasingly prioritising credential theft. Netskope said phishing campaigns now commonly use fake login pages, malicious open authorisation (OAuth) applications and brand impersonation techniques to bypass passwords and multi-factor authentication.
While cloud services were the primary overall target, banking credentials remained especially attractive because a single compromised account can provide access to broader systems.
The report also points to wider data security challenges for banks. Attempts by employees to share sensitive information with generative AI tools more than doubled over the past year.
On average, organisations recorded 223 monthly incidents of staff including regulated data, intellectual property, source code or passwords in AI prompts or uploads.
Only around half of organisations have data loss prevention measures in place to prevent sensitive information from leaking through AI applications.
The use of personal tools in the workplace continues to pose risks. Nearly one-third of employees uploaded data to personal cloud applications each month, and 60% of insider-related incidents involved personal cloud usage.
Almost a quarter of organisations lack real-time controls to detect or prevent data leaks via these platforms.
Netskope warned that these trends leave banks particularly vulnerable as attackers exploit trusted services and human error, adding that while phishing and data leakage threats are becoming more sophisticated, many organisations—including those in financial services—are still slow to implement effective safeguards.
Click here for more on Technology
