Sapia.ai launches world-first model to detect AI-generated content in online chat interviews
Sapia.ai, the world’s only smart hiring automation platform powered by deep-learning AI, has released a new function which will detect and flag responses sourced by generative AI, such as ChatGPT, in real time.
A world first, the new function draws from Sapia.ai’s growing proprietary dataset of over one billion words — collated from over 12 million responses from 2.5 million candidates that have used its platform.
Brands including Woolworths Group, Holland & Barrett, and WOLT trust Sapia.ai to accelerate and enhance their recruitment and promotion processes. A conversational, Natural Language Processing (NLP) based chat AI interviews, assesses and screens for the best talent at scale via an easy to use messaging platform.
In addition to improving diversity outcomes by eliminating unconscious bias, it also allows companies to re-allocate thousands of hours spent screening talent towards higher value tasks.
This new feature prevents candidates from using generative AI tools to respond to prompts from Sapia.ai’s platform. Candidates will be alerted in real-time as they respond when their answers are likely to be AI generated content (AGC), giving them an opportunity to change it ahead of the final submission. Failing this, it will then flag to the decision-maker the likely inclusion of AGC in the candidate’s response for further review.
Barb Hyman, Sapia.ai CEO and founder, said: “This is something our competitors can’t do. It’s our competitive moat. While it is possible to detect use of generative ChatGPT through analysis, we’re conducting it in real-time. Our data set also gives us the ability to readily adapt to new iterations of generative AI.”
“It builds on the plagiarism flag which we released in 2019. It identified candidates who sourced their answers from the internet or other candidates. Adding a new flag for AI generated content maintains the integrity of the structured interview as a fair, and accurate way to assess talent.“
“What is equally important to our product integrity is that we are transparent with the individual – transparent right from the outset that their interview responses are being assessed for AGC, and giving them the option real time to change their answers. We have always built our product with the human experience front of mind. We aim to put people at ease, having the interview blind and untimed and letting them know how their data is being used so they trust the experience.
“We believe every candidate matters. Today’s applicant is tomorrow’s customer. This is why we strive for exceptional candidate satisfaction metrics, with up to 94% completion rate for our long-standing customer Qantas Group, and an average completion rate of 91% across all our customers.
Sapia.ai’s Chief Data Scientist Dr. Buddhi Jayatilleke says the accuracy of the Sapia.ai’s AGC detector comes from the unique data asset the company can leverage to build and test the feature.
“We tested our AGC flag with thousands of generated answers from GPT-2, GPT-3 and ChatGPT on various prompts related to multiple role families. We were able to achieve a ROC-AUC of over 95%, which is a strong indicator of the accuracy of a classifier. This is due to our ability to distinguish the differences between human-written text and the formulaic nature of content coming from generative AI models, leveraging our large human-written response data set,” he said.
About Author
