Co-pilot now scans every uploaded resume for hidden content and, when it detects a potential attempt to manipulate the AI’s assessment, shows a warning on the candidate profile.
The warning gives you the full picture: the hidden text itself, how it was concealed — such as text matching the background color, tiny fonts, or transparent text — and why it was identified as a potential manipulation attempt.
A shield badge on the resume also shows the scan result at a glance: purple when no hidden instructions were detected, and red with a count when hidden instructions were found and ignored.
Detected hidden instructions are ignored by Co-pilot, and candidates are never scored down or rejected automatically because of a detection — the decision always stays with you.
Every detection is recorded in your audit log, and you can turn the warnings off anytime under Settings → Co-pilot.
