First published on LinkedIn
Video interviewing has evolved to become an essential component in talent acquisition. Now that every smartphone and tablet has a built-in camera — and candidates are largely comfortable with video calling via apps such as FaceTime and Skype — every recruitment team needs to understand that video interviews have changed: Video assessment using AI technology is the future and will make your hiring process more efficient and effective.
Video assessment using AI technology is the future and will make your hiring process more efficient and effective.
AI-enabled video assessments offer several advantages when used as part of the hiring process. The primary value of these types of video assessments in talent assessment comes from using facial recognition technology to verify that the person doing the interview or taking an assessment is actually the candidate.
Video assessments can also be used to analyze the content of the video interview using natural language classification and processing. This practice is supported by research, can reduce unconscious biases and helps improve hiring outcomes.
Here’s how employers can avoid the problems of video talent assessments while maximizing the benefits.
Some AI vendors offer expression recognition, but that technology presents many potential data pitfalls and legal challenges, and it opens up the possibility of bias influencing the assessment. Expression recognition isn’t appropriate for identifying stable traits — for example, just because a candidate looks happy doesn’t mean they’re friendly.
In more specific cases, the use of appearance or expressions can introduce significant biases and even result in unfair treatment. A candidate may have a facial tic or some medical condition that limits their ability to control their expressions. For example, a person that has suffered a stroke may have limited use of their facial muscles, but facial recognition software would not be able to take that context into account and might interpret their behavior negatively as a result, leading to a poor assessment of the candidate. That candidate might be penalized in the assessment process as a direct result of a medical condition, creating the potential for discrimination and unfair assessment practices.
In short, expressions shouldn’t be analyzed to reward or penalize a candidate because they don’t necessarily have anything to do with the traits needed to succeed on the job.
By contrast, focusing on the content of a candidate’s response is a big step forward for recruiters. Natural language classification and processing can clean up unstructured data from audio files in a comprehensive manner, resulting in a written transcript that can be scored by the AI.
Traditional assessment processes have shown that what people say about their past behavior in specific real-world situations is a reliable predictor for future behavior and for their traits and skills. This holds true especially for structured interviews, which should be the core of the video-interviewing process.
The AI looks at this text in a way that removes any unconscious human bias while searching for evidence of good job-related behavior. An AI doesn’t make human errors because of bias or tiredness. This objectivity means that different people with the same response receive identical results.
The AI looks at this text in a way that removes any unconscious human bias while searching for evidence of good job-related behavior. An AI doesn’t make human errors because of bias or tiredness.
Ultimately, focusing on what people actually say, not on how they look, in video interviews leads to a more defensible and transparent assessment process.
Watch our on-demand webinar on AI in video interviewing to learn more about how speech to text is the robust unbiased future of video assessment.
About the AuthorFollow on Linkedin More Content by Evan Theys