A study by recruitment agency Oriel Partners reveals that ChatGPT lied on average 14 times on CVs to help candidates improve their chances of getting interviews.
The primary aim of this research was to assess the extent to which ChatGPT can modify and enhance CVs and to understand the implications of these changes on the recruitment process. The study’s findings are particularly significant in an era where AI tools like ChatGPT are becoming increasingly sophisticated and accessible.
Oriel Partners embarked on a meticulous research process, employing ChatGPT to enhance 100 real CVs submitted for a specific job listing. The study compared these AI-enhanced CVs against the original human-generated CVs to identify the extent and nature of modifications made by the AI.
A custom GPT-powered tool was developed to screen both sets of CVs, assigning scores based on their alignment with the job description. This approach allowed for a direct comparison of the effectiveness of AI enhancements in improving a candidate’s prospects.
The study revealed that ChatGPT made an average of 14 embellishments per CV. These enhancements ranged from subtle tweaks in phrasing to significant additions in skills and experiences, raising concerns about the authenticity of AI-enhanced CVs.
Embellishments to CVs | Avg. Number of Embellishments |
“Embellishments” to Profile section | 7 |
“Embellishments” to Key Skills & Attributes | 4 |
“Embellishments” to Professional Experience | 3 |
Total | 14 |
AI-enhanced CVs scored significantly higher on average compared to the original CVs, suggesting that these embellishments effectively improved candidates’ chances of securing interviews. However, this also highlighted the potential for AI to create an uneven playing field, where candidates using AI might have an unfair advantage.
Type of CV | Avg. Scores |
Embellished CVs Avg. Score | 9.4 |
Normal CVs Avg. Score | 8.3 |
In light of these findings, and considering that a recent survey by Kaspersky showed 42% of workers would use AI tools like ChatGPT to refine their job applications, it’s clear that AI’s role in recruitment is expanding rapidly. This highlights the importance for employers to create methods to detect and lessen the effects of resumes created by artificial intelligence.
Olivia Coughtrie, Co-Founder at Oriel Partners commented: “It’s extremely concerning that ChatGPT is able to lie and create complete fabrications on CVs. This will make it virtually impossible for employers to screen genuine applicants, meaning more rigorous work will need to be carried out during the interview process to determine the best candidates.”
Oriel Partners recommends that employers develop strategies to identify and mitigate the impact of AI-enhanced CVs, such as incorporating more in-depth interviews and skill assessments. For job seekers, the study serves as a reminder of the importance of authenticity in their applications.
This study is a crucial step towards understanding and managing the role of AI in recruitment. Oriel Partners calls for a balanced approach to AI use, advocating for responsible and ethical practices that benefit both employers and job seekers.