Catch ChatGPT-Generated Code in Coding Interviews
1 in 10 developers now cheat on tech assessments with ChatGPT or Copilot. ClarityHire's code coherence AI, keystroke biometrics, and paste detection catch it — with evidence you can verify.
Code coherence analysis
Multi-layered detection
Evidence on every flag
Browser-native, candidate-friendly
Multi-signal detection of AI-generated code
Not just plagiarism detection — ClarityHire scores the edit pattern, not just the output.
Code coherence AI
Claude analyzes the edit pattern of every session. LLM-generated code gets flagged with explainable reasons.
Keystroke biometrics
XGBoost classifier detects typing-rhythm anomalies that indicate another person or an LLM took over.
Paste event logging
Every paste is logged with source and code-size correlation — suspicious bulk pastes get flagged.
Gaze anomaly detection
Frequent off-screen glances during coding tasks suggest candidates are reading from another screen.
Face continuity
ArcFace-based face matching confirms the same person throughout the assessment.
Tab-switch tracking
Frequency and duration of tab switches correlate with off-platform LLM use.
Edit pattern analysis
Authentic coding shows iteration — type, test, refactor. LLM code arrives fully formed; that difference is a signal.
Per-signal authenticity score
Each session gets a score per signal so you see exactly where cheating appeared — no black-box verdicts.
Why teams pick ClarityHire for this
Real workflow, built-in integrity, end-to-end pipeline — in one platform.
Black-box detection creates false accusations
Some tools output a cheating verdict with no evidence. ClarityHire gives you the edit pattern, paste log, and AI reasoning behind every flag — so you can verify before acting.
- Natural-language AI explanation per flag
- Evidence links to session replay
- Adjustable thresholds per role
Every signal, every session
Keystroke biometrics, face continuity, gaze analysis, and code coherence AI run silently throughout every assessment. Every signal feeds a per-candidate authenticity score you can review or automate on.
- Ten signal types including keystroke, gaze, face, paste events
- Per-signal authenticity score per submission
- Human-readable explanations on every flag
One platform, not five tools
Job boards, assessments, live video interviews, candidate tracking, structured feedback, and PDF reports — all in ClarityHire. Stop paying for five vendors.
- Replaces ATS + assessment + interview + proctoring
- Kanban pipelines with stage automations
- PDF hiring reports for stakeholders
How it works
Enable integrity
Set integrity level (basic/strict) per assessment. No candidate-side installs required.
Candidates code
Every keystroke, paste, edit, and camera frame captured in the background.
AI scores the session
Code coherence AI plus signal aggregation runs on submission. Authenticity score + flags generated.
You verify flags
Review replay + explanation for every flag. Decide whether to proceed, follow up, or reject.
Frequently asked questions
How does code coherence AI actually detect ChatGPT?+
ChatGPT-generated code typically arrives as large clean pastes with consistent style and no iterative debugging pattern. Claude analyzes the edit timeline (type → test → refactor) and flags patterns inconsistent with authentic human coding. Every flag includes a written explanation.
How is this different from plagiarism checkers?+
Plagiarism checkers compare against a known corpus. Code coherence AI analyzes the edit pattern — how the code was written — not where it came from. It works even for freshly-generated LLM code that has never been seen before.
What if a candidate genuinely writes clean code quickly?+
Fast clean coding still shows iteration and style variation. Code coherence AI looks at the full session trajectory, not a single snapshot. Confidence scores are adjustable per role so you can tune for senior candidates.
Will candidates know they're being monitored?+
Yes. Candidate transparency matters — they see a session recorder indicator and review consent terms on start. Most don't notice detection is running; it's all browser-native with no extensions.
Can I disable AI detection for specific roles?+
Yes. Integrity runs at three levels: none, basic (keystroke + paste + face), and strict (all ten signals). Dial per assessment based on role sensitivity.
Explore related features, roles, and use cases
ClarityHire is one platform. Every feature is built to work with the rest.
Code Coherence AI
Claude-powered analysis of edit patterns catches LLM-generated code.
Cheat Detection
Ten signal types including keystroke, gaze, face, and paste.
Keystroke Biometrics
XGBoost classifier detects takeovers and proxy test-takers.
Take-Home Coding Tests
Async take-homes with AI-proof integrity.
AI-Proof Coding Tests
Deep dive on AI-proof assessment design.
Catch AI-Generated Code Without False Accusations
Turn on code coherence AI in one click. Explained flags, human in the loop, always.