Anti-cheating encompasses the full range of technical and pedagogical measures designed to detect or deter academic dishonesty in online assessments. On the technical side, the main tools are lockdown browsers that restrict the test environment, focus-loss detection that flags when a learner switches away from the exam window, similarity detection that compares submitted text against reference corpora, and AI-based anomaly detection that looks for unusual response patterns. Pedagogical deterrence — using unique, scenario-based, or open-ended questions that resist simple copying — is often more effective and less invasive than technical countermeasures alone. No single measure is foolproof: lockdown browsers can be bypassed with a second device; focus-loss signals produce false positives when a learner uses assistive technology; and similarity detectors flag coincidental matches rather than proven intent. The most defensible approach layers multiple light signals rather than relying on one decisive flag, and always routes ambiguous cases to a human reviewer before any penalty is applied. Assessment integrity policy must also be transparent to learners so they understand what is monitored and what consequences a flag triggers. Keeping this policy visible in the exam interface — not only in a separate terms document — is a simple design decision that reduces disputes after the fact.
