More than half of the world’s mobile banking applications — 1,384 out of 2,686 examined, contain at least one high or critical security flaw, according to a new assessment by security firm Ostorlab.
The audit covered apps from 487 banks across 64 countries, spanning Android, iOS, and Huawei’s HarmonyOS. It did not test live systems or confirm active exploits, but identified concrete weaknesses that could enable account takeover, credential theft, or fraud if exploited.
Biometric authentication is especially fragile on Android. Ostorlab’s scanner successfully bypassed fingerprint and facial recognition flows in 35.1% of Android banking apps tested, even though 73.2% of eligible apps tied biometric verification to cryptographic keys. That gap suggests implementation flaws, not just theoretical risks.
Hardcoded secrets were another major red flag. Private keys, cloud credentials, and service access tokens appeared inside app packages in 20.3% of apps (544 total). Google API keys showed up in 72.5%, not inherently dangerous, but risky when misconfigured or unrestricted, potentially enabling abuse of Firebase, Maps, or other cloud services.
Outdated software components compound the problem. 41.1% of apps included at least one third-party library with known vulnerabilities. 18.2% used libraries carrying critical-rated flaws. And 99.5% of Android banking apps still support OS versions no longer receiving security updates, leaving users on older devices exposed to unpatched platform-level bugs.
Modern protections remain rare. Only 4.9% of apps support passkeys, a phishing-resistant alternative to passwords. Just 20.2% use application or device attestation, which helps banks verify they’re talking to a genuine, unmodified app, not malware masquerading as one. The remaining 79.8% offer no such defense.
Worse, 22.3% of Android banking apps can detect remote-control or screen-sharing tools, commonly abused in real-time banking fraud scams where attackers guide victims through transactions. Application shielding, software designed to resist tampering, appeared in only 26.9% of Android apps.
Apps from 192 of the 500 banks studied also appear on public lists of institutions targeted by known banking malware. That doesn’t mean those apps have been breached, but it signals active interest from threat actors.
Ostorlab stresses its findings are evidence of presence, not proof of exploitability. Still, the scale is striking: 51.5% with high or critical flaws, 18.4% with at least one critical-only finding, and nearly all apps running on obsolete OS foundations.
No regional breakdowns were provided for India or the US specifically, but the sample includes banks serving both markets, and the flaws apply uniformly across platforms. Customers in both countries rely heavily on mobile banking, making the implications direct.
Banks now face pressure to patch, update, and adopt stronger safeguards, not just in code, but in how apps are built, tested, and distributed. Regulators in India (RBI) and the US (CFPB, OCC) have issued guidance on mobile app security, but enforcement remains uneven.
Ostorlab’s report is live. No coordinated disclosure timeline or vendor-specific remediation status was published with the findings.
SDFiled by The Software Desk
Apps, platforms, operating system releases, security fixes and the features arriving in them.
More from this desk →
Be the first to comment
Join the argument. No password, just your email or a passkey.