We’re committed to maintaining the integrity of our structured chat interview as a fair and accurate way to assess talent. As a written interview used to analyze personality and determine role fit, it's essential to ensure that candidates submit answers they wrote themselves.

Our team at Sapia Labs has developed a world-first detector that analyzes candidates' interview responses and flags if there is a high likelihood that they have been generated with AI such as ChatGPT.

Customers can choose the threshold at which an interview is flagged. The default behavior is that a warning flag is generated for candidates and hiring teams if artificial content is detected in all five free-text questions; however, this can be customized to any number of questions from September onwards.

Candidate experience

A candidate will receive a pop-up warning if the detector flags their answers for the number of responses set for that customer. It will tell the candidate that artificially generated content has been detected, and give them a chance to reconsider their answers or submit as-is.

Hiring team experience

If a candidate has been flagged, the flag warning will show up on Edge or in the ATS, with the details visible on the Talent Insights profile, letting hiring teams know that artificially generated content has been detected, and which answers have been flagged.