Common warning signs include unexpected pop-up login screens, repeated permission prompts, disabled security protections, unusual SMS activity, and apps requesting accessibility or device-admin style access without a clear business reason. Malware may also stay dormant until the device moves or becomes active, so suspicious behaviour can appear only after normal use rather than at installation time.
What a live credential-harvesting infection looks like on a banking phone
An active banking trojan is usually trying to collect something immediately useful, not just sit on the device. The strongest clue is a change in how the phone behaves during banking, such as a fake overlay appearing at the moment a real app opens, a sudden request for accessibility access, or SMS-based prompts that do not match normal banking flow.
When those behaviours line up with unusual background activity, the infection is often trying to steal credentials, intercept one-time codes, or watch for the moment the user enters a bank username and password. The practical signal is not one symptom in isolation, but a cluster that appears during app use or shortly after the device becomes active.
Mobile banking trojans often rely on overlays, accessibility abuse and SMS interception to capture login data and verification codes, which is why unexpected prompts and screen interference matter so much. A useful reference point is Android Banking Trojans Guide, which explains the main fraud paths and the controls banks use to detect them.
Which warning signs are most consistent with active harvesting?
The most telling signs are user-facing and timing-based. Unexpected pop-up login screens, repeated permission prompts, disabled security protections, unusual SMS activity, and apps requesting accessibility or device-admin style access without a clear business reason all suggest the malware is trying to reach sensitive input or bypass normal phone controls.
Another useful clue is dormancy that ends when the device is in normal use. Some malware avoids attention until the user opens the bank app, unlocks the screen, moves the device, or receives an SMS, so a phone can look clean at install time and still be compromised later. That delay is itself suspicious when combined with credential-entry attempts or code interception.
For mobile app abuse patterns more broadly, Mobile Malware Guide is a useful companion because it maps the common banking trojan behaviours, including overlay attacks, accessibility abuse and OTP theft, to practical mobile detection and control points.
How to separate suspicious behaviour from ordinary app friction
Not every prompt means infection. Some banking apps legitimately ask for permissions, device checks, or notification access. The difference is whether the request is expected, consistent with the app’s stated purpose, and proportionate to the task. A banking app that suddenly wants accessibility access, admin rights, or SMS visibility without a clear explanation should be treated as high risk.
Context matters as much as the alert itself. A single pop-up is weaker evidence than a pop-up that appears only when the banking app launches, or a permission prompt that reappears after being denied. The more the behaviour depends on banking activity, the more likely it is that the malware is waiting to capture credentials or verification codes rather than causing generic device instability.
If the warning signs are tied to a specific app, compare that app’s permissions and installation source against its expected function. OWASP Mobile Top 10 M1 Improper Credential Usage Guide helps frame whether a mobile app is exposing credentials or using weaker storage and authentication patterns than it should.
Risk and Threat Considerations
A phone that is harvesting banking credentials can expose both the login secret and the second factor path. That makes the infection more dangerous than a generic nuisance app, because the attacker may capture passwords, session material, or SMS codes in the same interaction window and then use them before the victim notices.
Failure mechanism: The malware waits for banking activity, overlays a fake interface, or abuses permissions such as accessibility and SMS access to read or relay the user’s credentials and verification codes.
Impact: The attacker can perform account takeover, drain funds, change account settings, or use the stolen access to pivot into other services that reuse the same credentials or device trust.
For credential abuse patterns and response logic, MITRE ATT&CK T1598 Phishing for Information is useful because it connects harvesting behaviour to the broader detection problem of credential collection by deceptive interaction. If the banking app is itself the target interface, OWASP API Security Top 10 also helps frame the back-end risk when stolen credentials are used against account and session endpoints.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
OWASP Non-Human Identity Top 10 and OWASP API Security Top 10 address the attack and risk surface, while CIS Controls v8 and NIST SP 800-53 Rev 5 set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| OWASP Non-Human Identity Top 10 | NHI-02 — Secret Leakage | Mobile credential harvesting often exposes reusable secrets and codes. |
| NHI-04 — Insecure Authentication | Fake login overlays and code interception weaken the authentication path. | |
| NHI-07 — Long-Lived Secrets | Stolen banking secrets become more damaging when they remain valid too long. | |
| Recommendation — Reduce exposed secrets and rotate any credentials the infection may have captured. Harden mobile authentication flows against overlays and code interception. Shorten credential lifetimes and revoke reusable secrets quickly after suspected harvest. | ||
| OWASP API Security Top 10 | API2 — Broken Authentication | Stolen credentials can be replayed against banking APIs and sessions. |
| API5 — Broken Function Level Authorization | Compromised logins may let attackers invoke privileged banking actions. | |
| API6 — Unrestricted Access to Sensitive Business Flows | Harvested credentials can drive fraud flows such as transfers or payee changes. | |
| Recommendation — Harden authentication checks and invalidate suspicious sessions quickly. Verify privileged banking actions are blocked unless the caller is explicitly authorised. Protect high-risk banking flows with step-up controls and abuse detection. | ||
| CIS Controls v8 | CIS-5 — Account Management | Credential theft requires rapid account review, revocation and recovery. |
| CIS-7 — Continuous Vulnerability Management | Malicious mobile apps and abused permissions need timely detection and removal. | |
| Recommendation — Review, revoke and recover affected accounts and access paths immediately. Continuously identify and remove malicious or vulnerable mobile software. | ||
| NIST SP 800-53 Rev 5 | IA-5 — Authenticator Management | Harvested passwords and codes require authenticator lifecycle control. |
| AU-6 — Audit Record Review, Analysis, and Reporting | Suspicious login prompts and SMS anomalies require log review and correlation. | |
| Recommendation — Rotate or revoke compromised authenticators and secrets promptly. Correlate device and account logs to confirm suspicious authentication activity. | ||
Practitioner Guidance
What to verify: Treat the device as suspicious if the same app repeatedly requests permissions, displays mismatched login screens, or starts generating SMS and accessibility prompts only when banking activity begins. Verify whether the banking app was installed from an official source and whether any new accessibility or device-admin grant appeared without a clear business reason.
What to prioritise: If a credential harvest is plausible, prioritise account protection over device cleanup. Freeze or monitor banking sessions, rotate the affected credentials where possible, and validate whether SMS-based second factors may already have been exposed or replayed.
Practitioner takeaway: The key question is not whether the phone looks generally infected, but whether the malware can see, overlay, or relay the bank login path in real time. If it can, assume the account is at risk even before you prove full device compromise.
Related resources from NHI Mgmt Group
Deepen Your Knowledge
Free weekly newsletter
Subscribe to the NHI & AI Identity Journal
The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.
Bonus 33% off our NHI Course when you subscribe.
Reviewed and updated by the NHIMG editorial team on October 31, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org