Last December 7, HR Intelligence participated in the BIAS project’s International Co-Creation Workshop in Venice in a wonderful setting of artistic beauty and enthusiastic, participatory people from all over Europe.
The BIAS project, part of the Horizon research program, funded by the European Union, aims to create an AI (Artificial Intelligence)-based tool – in technical terms a debiaser – for the mitigation of bias (i.e., subjective distortions of reality based on prejudices and stereotypes).
Specifically, the event gave us an insight into how Artificial Intelligence can be used in recruiting processes, what critical issues may arise, and what might be the tools to overcome them.
The use of AI in recruiting processes
AI used in recruiting processes is based on natural language processing (NLP), which can analyze text to draw inferences or decisions.
However, NLP-based systems are not free from implicit biases, which are acquired from the models they are based on and also contain stereotypes and biases of our society.
These biases can have a strong influence on the decision-making process and lead the selector(s) to make choices based on biases and stereotypes.
During the day, we used AI-based systems for recruitment and selection and, in working groups, were able to experiment with and comment on the results provided by the first debiaser prototype on two distinct topics:
- the comparative evaluation of job applicants, starting with CVs and cover letters;
- the highlighting of some potential biases and suggested mitigation measures derived from the analysis of CVs and cover letters.
We shared experiences and best practices, emphasizing the importance of implementing balanced and responsible AI policies within organizations.
Then, in plenary, we addressed issues regarding the reliability and correctness of debiaser results and explored the needs and expectations of the different stakeholders present, which included, for example: AI experts, HR professionals, researchers, and members of the public administration.
We helped bring to life content rich in different perspectives and elements, highlighting the importance of considering ethical and social implications in the design of algorithms and advanced technologies from various perspectives.
Debiasing AI: the tool to eliminate Bias in recruiting processes
The debiaser tool being developed not only promises to improve the neutrality of AI decisions, but also to lay the groundwork for a more ethical and responsible approach to technology implementation.
The workshop highlighted the importance of considering social, cultural and ethical dimensions in algorithm design, promoting an inclusive and informed approach to emerging AI-related challenges.
The BIAS project will continue, to develop and implement innovative solutions for debiasing in artificial intelligence.
The Venice event demonstrated that international collaboration and engagement of diverse perspectives are critical to addressing the complex challenges associated with AI and shaping a more equitable and sustainable future.
Add a Comment