A few days ago, I participated in a workshop given by Ms. Sandrine Théard, on artificial intelligence and its benefits in the coveted staffing process. I embarked on this training session very curious about what I could learn, but also skeptical on the edges, because although the use of AI is inevitable in our professional lives, is its use regulated? What about IA for staffing?
In this workshop, we focused on generative AI, a branch of artificial intelligence that focuses on the creation of content (data, images, texts, sound, etc.), including the famous GPT Chat.
But how can generative AI transform recruitment?
There are lots of ways to use AI in our process, the idea is to capitalize on non-value-added tasks, which a human can avoid. Here are examples related to two (2) important stages of the process where AI becomes our best ally.
In a highly competitive market with a shortage of talent, AI is proving very useful when it comes to stimulating creativity for the creation of job postings, for example. Indeed, it is possible to work on an original display, by feeding the “machine” with criteria from your persona of the ideal profile and elements of your context (sector of activity, industry, location, etc.). It is even possible to achieve inclusiveness by adding an epicene writing criterion and the omission of discriminating terms.
This is one example among so many possibilities, but you will already be innovating with your traditional tools.
Analysis of applications
The arrival of ATS in the world of recruitment has made our sorting of resumes much easier, meaning matching the requirements of the position and the professional background with the skills mentioned by the candidates. In addition, with ATS like those in Indeed or the SmartRecruters, the management of applicant tracking has improved greatly and has been centralized. However, if there is a volume to be assessed, how can you be more efficient and save time? AI can be used to analyze and filter in order to bring out profiles corresponding to the requirements of the position to be filled. Sorting is therefore done automatically using filtering algorithms and this contributes to considerable gains in efficiency and time!
Should we be wary of AI?
Obviously, AI is not an infused science (no pun intended) and has its limitations that must be taken seriously. It is especially important to know how to use the machine, use it wisely and not use it if you do not master the subject for which you are asking for help. During our training, Sandrine told us that we must treat AI like a trainee, it can be useful, but we remain the expert to validate its work. Here are some important limitations to note:
- The desired resume filtering algorithms is only effective in positions with technical skills, unlike positions where human skills are emphasized.
- There is an ethical issue, because our responsibilities, as HR professionals, must not be transferred to AI.
- It may result in the standardization of profiles during automated sorting of resumes if AI is poorly used. This can encourage the infiltration of several biases.
- There is what we call “deep fake”. As much as AI will cross-reference data and bring out very interesting profiles according to your criteria, job seekers also use AI to perfect their image.
What did I think of this workshop in general?
Very happy with the learning, both the advantages and the limitations of AI. Concrete examples were shared with us, but above all we were made aware to the diligent use of AI for the purposes of our staffing processes. Before you get started, consider and evaluate your choices of suppliers or AI applications. Above all, do it while respecting your context, because this will guarantee better integration and the expected results.
By Noellie Dias, CRHA – Collaborator to the blog