Sapia Labs

The science behind our ethical AI

How our AI Smart Interviewer achieves unprecedented results

Why

Interview by chat works

Language defines who we are – it’s what makes us human. It encodes deep signals that reveal our personality, our skills, and our experiences. Looking at it through the lens of personality theory, lexical analysis of language helps us to discover and understand personality traits. And we get better at it all the time.

In AI job interviews, we gain insights through language. However, face-to-face interviews are flawed: In person, we deduce and infer and conclude things about each other through visual and audial signals. That is where biases, those both conscious and unconscious, come into play. You might think you know someone, but you may just be looking in a funhouse mirror.

 

The best

Of both worlds

Our Smart Interviewer combines the benefits of the traditional job interview with the latest in NLP (Natural Language Processing) and machine learning, to enable fairer and more objective analysis of interview responses.

Our Smart Interviewer uses:

  1. Structured interviews, where the same questions are asked of every candidate in a controlled conversation flow, and evaluated using a well-defined scoring rubric. Structured interviews have proven to be one of the most reliably predictive selection methods concerning future job performance, and they significantly reduce bias when compared to unstructured interviews.
  2. A blind text-chat based interview. By ingesting only the text responses as fed to our interview AI, the evaluation remains free of bias-inducing visual and audial signals.
  3. Peer-reviewed research, and the latest advances in NLP, machine learning, and optimization algorithms. We are able to extract the language signals that reliably predict job performance, such as personality traits, behavioral competencies, and language skills. In doing so, our interview AI can provide an unbiased assessment of each candidate’s interview responses.

Backed by science

Browse summaries of our latest research

Blog

Question-aware outlier answer detection for fairer AI scoring of interviews

Artificial Intelligence-based interview scoring learns from past interview answers, which makes it hard for it to determine if a candidate is legitimately answering the question if their response incl Read More
Blog

Plagiarism detection and prevalence in online, text-based, structured interviews

Faking is a common issue with traditional self-report assessments in personnel selection (Levashina et al., 2014). The major co Read More
Blog

Identifying and mitigating ethnicity bias in structured interview responses

Discrimination based on race and ethnicity in personnel selection is a well known and pervasive issue highlighted in numerous studies (Bertrand & Mullainathan, 2004; Kline et al., 2021; Pager et a Read More
Blog

Reducing disability bias with chat-based interviews

Bias and discrimination against candidates and employees with disabilities continues to be an increasingly important topic 30 years after the Americans with Disabilitie Read More

Browse our other research and whitepapers here

Research and whitepapers
Blog

Elevating Candidate Experience: Empowering Candidates With Language Choice

Read More
Blog

How blind scoring of Video Interviews improved hiring diversity by 20%

Read More
Blog

The Changing Role of the Organizational Psychologist As HR Embraces AI

Read More

Sign up to our newsletter

If you’re ready for change, we can make it happen

Book Now