Does TestGorilla detect cheating?
By Aaron Cao · Updated
Yes, within limits. TestGorilla can take periodic webcam snapshots, detect when you leave full screen or switch away, log your IP address and device, and draw questions from randomized pools. It reports these signals to the employer; a person decides what they mean.
The anti-cheating measures TestGorilla actually runs
You have an assessment link in your inbox and you want to know what it can observe before you click it. This section names the specific checks, so you can prepare instead of speculate. They fall into four groups, and the employer chooses which ones apply.
- Webcam snapshots. Still images are captured at intervals during the test rather than a continuous recording, and they appear in the employer's report.
- Full-screen and focus detection. Leaving full screen, switching windows, or moving away from the assessment is logged.
- IP address and device logging. The location and device used are recorded, which surfaces a test taken somewhere unexpected.
- Randomized question pools. Questions are drawn from a larger bank, so shared answer sheets from a previous candidate are a poor investment.
Time limits per question do similar work. A tight clock leaves little room to look something up without the timer showing it.
What the snapshots and flags mean to an employer
A flag is data, not a verdict. The report shows an employer that you left full screen twice, or that a snapshot was taken with nobody in frame. What happens next depends entirely on that company's process and on how the rest of your application reads.
Consider a concrete case. A customer support candidate takes a TestGorilla typing and situational-judgement battery from a coworking space with unreliable wifi, reloads the page after a drop, and the report shows two focus losses. The hiring manager sees a strong score, an ordinary explanation, and moves on. The same two flags next to a borderline score read very differently. Context does the deciding.
The broader question of what interview software can and cannot observe is covered on the detectability topic hub.
Where the monitoring stops
Browser-based monitoring only sees the browser and, when enabled, the camera. It does not inspect other applications on your computer, a second monitor, a phone next to the keyboard, or another person in the room but outside the frame. TestGorilla does not put a human proctor on a live video feed watching you work.
That boundary cuts both ways. It means the checks miss things, and it means the checks flag innocent behaviour: a notification stealing focus, a browser crash, a shared home IP address. Employers who understand this treat the report as one input rather than a filter, which is why a live conversation almost always follows a passing score.
Where SubcueAI fits, and where it does not
SubcueAI assists live, spoken interviews. The native macOS and Windows app captures system audio and your microphone and shows suggestions in a floating local overlay; the browser extension's Side Panel does the same for a call running in a Chromium tab, capturing the meeting tab's audio only. No meeting bot joins, and nothing is injected into the meeting page.
A TestGorilla assessment is a typed, in-browser test, often with a camera watching. SubcueAI does not read or answer it, and an assessment with webcam monitoring enabled is out of scope. Aaron Cao, founder of SubcueAI, drew that line deliberately: a product that claims to beat proctoring is making a promise it cannot keep on someone else's computer.
The stage worth preparing for is the interview the assessment unlocks. A timed run on the mock interview page rehearses saying your answers out loud, which is the skill that round tests.