TestGorilla can detect some behaviors associated with using outside help, but it does not automatically prove that a candidate used artificial intelligence. Its assessment tools look for activity signals such as tab switching, leaving full-screen mode, pasting content, and other behaviors configured by the employer.
What triggers a cheating flag on TestGorilla?
TestGorilla can flag activity such as switching browser tabs, leaving full-screen mode, opening developer tools, pasting content into answers, or disabling the webcam. A flag records a behavior; it does not prove that a candidate cheated or used ChatGPT.
TestGorilla's Help Center lists these monitored signals:
- Tab switches: moving to another browser tab during the assessment.
- Full-screen exits: leaving the full-screen test window.
- Developer tools: opening the browser's developer console or inspection tools.
- Pasted content: when and where text or code is pasted into an answer.
- Webcam status: whether the camera is enabled or turned off, where camera monitoring is required.
The platform also captures webcam snapshots when enabled and records IP-based location information. These provide context for reviewing identity and possible repeated attempts; they are not proof of cheating by themselves.
What does an orange or red flag mean?
TestGorilla groups behavior into No Issues, Some Issues, and Major Issues. Its guidance gives moderate tab switching or full-screen exits as examples of Some Issues, and frequent full-screen exits or a disabled webcam as examples of Major Issues. The overall tier considers the pattern of activity rather than treating every event as equally serious.
Can the employer see the cheating flag?
Yes. Employers can review the behavior monitor, flagged-event counts, and the behavior log alongside assessment results. Webcam snapshots can also be reviewed when that feature is enabled.
A flag does not automatically mean rejection. TestGorilla advises employers against rejecting a candidate solely on anti-cheating signals. An accidental tab switch, a system notification, or a technical issue may explain an event. If something unexpected happened, tell the recruiter what occurred so they can review it in context.
Can TestGorilla tell if you used ChatGPT?
Not with certainty from the answer alone. A platform may flag behavior that suggests a candidate left the assessment, pasted text, or used another application, but a behavior flag is not the same as proof of AI authorship.
The employer typically reviews the assessment result and any available monitoring information together. A single event may have an innocent explanation—for example, an accidental tab change or a paste from permitted working notes—so candidates should follow the instructions supplied with the assessment.
Does TestGorilla record your screen?
TestGorilla's anti-cheating controls are not the same as continuous screen recording. Activity monitoring can record events or signals, while camera features may capture identity snapshots or video answers if the employer has enabled them.
Detection approaches differ across platforms — for example, does TestDome detect cheating through webcam snapshots, screen capture, and AI-video analysis rather than a single check.
How to avoid an accidental flag
Before starting the assessment:
- Read the employer's instructions carefully and confirm what resources are allowed.
- Close unrelated tabs, messaging apps, and browser extensions.
- Keep your phone and other reference material away unless the instructions permit them.
- Avoid copying and pasting unless the assessment explicitly allows it.
- Test your webcam and connection if camera access is required.
If you need an accommodation or the instructions are unclear, ask the recruiter before beginning rather than guessing.
If a proctored assessment doesn't go well, the outcome depends on the employer — can you still get hired if you fail an assessment test explains how employers actually use the results.
Bottom line
TestGorilla can monitor behaviors that may be associated with outside assistance, including some activity that could occur while using AI tools. It should not be treated as an infallible detector of AI-written answers. Follow the assessment rules, keep your testing environment clean, and ask the employer for clarification when a monitoring requirement is unclear.