Jayne Kettles is the co-founder and chief product officer at GR8 People. She’s been a technologist, product visionary, and innovator in the enterprise software industry for more than 15 years. A senior executive with strong leadership skills and technical expertise, Jayne has a history of delivering value to customers through innovative software and technology. In this blog she shares thoughts on the challenges and opportunities AI brings to talent acquisition.
AI and recruiting. These two words put together have generated a lot of buzz in talent acquisition. Adding “bias” into that mix has also created a different kind of buzz. What are your thoughts on inherent causes of process-enforced bias that’s common in modern recruiting systems?
Bias in recruiting can be caused by a number of factors, some totally natural to the human condition, some caused by prejudice, and others an outcome of weighing the benefits of process standardization and efficiency over fairness.
To achieve efficiency, efforts are made to identify repeatable patterns. One of the most common, and most associated with recruiting efficiency, is the “ideal candidate profile.” This may be a combination of a number of factors such as where a job seeker went to school and what they studied, or where they worked previously and in what roles, and much more.
The identification of an ideal candidate profile to quickly find and court only the most likely-to-perform talent reduces cycle time, cost-to-hire and better outcomes. However, the patterns can potentially open the door to bias.
Organizations try to develop their hiring patterns based on the study of their existing high performing and high value employees. If that pattern is based on male or female dominated roles, the ideal candidate profile may reinforce that bias. This may be a basic example, but in reality, biased outcomes have serious consequences. Compliance is critically important, but so too is the less-visible but still problematic impact of not having enough diversity of thought and experience that drives innovation and a competitive edge.
Share your thoughts on how companies will need to address inherent bias in data used along with AI/machine learning for sourcing, matching and scoring candidates?
Purposeful design, ongoing visibility and diligence!
The minimization or elimination of bias needs to be made a goal for the system and the AI tools we wrap around the process.
Because this is so fundamental to the proper application of these automated technologies, best-in-class AI providers across the whole spectrum of use-cases are training their tools to correct for bias prior to putting them in the hands of recruiting organizations. To this point, the forward-thinking administrators of these tools will not only be asking IF the tool considers and corrects for bias, but also WHICH biases it measures and HOW as well. You can spot a quality provider if the answer seems well-considered and thorough, with data that can illustrate outcomes.
As anyone who has been bitten by overhyped technology can attest, there is a big difference between understanding what a system can do and ensuring that it is having the desired effect against your real-world scenario.
Does AI offer us the opportunity to potentially remove bias from hiring?
When used consistently and when developed in a considerate manner, properly-trained AI can automatically adjust itself to ensure that its outcomes are not generating bias. Right now, however, AI is not solely responsible for hiring. In most cases it is being used to discover, sort, rank and prequalify talent in as non-biased a way as possible before providing that list to the human-driven portions of the selection process.
At the end of the day (today) a human is still typically going to be interviewing a person and deciding if they should get the job. Machines may be able to give a confidence rating to submitted candidates, but if the machine-championed candidate can be rejected by a hiring manager for a thousand other human bias drivers, bias can be reintroduced.
Evaluating today’s landscape, what advice can you offer someone in talent acquisition who wants to choose the right solution?
With all solutions in this space in particular, be wary of fantastic claims that are unsupported by large sets of data. We have seen many providers claim that their point solution can replace swaths of existing processes and solutions, only to look a little deeper and see that the scientific validity, architectural underpinnings, and/or basic enterprise-readiness of the vendor does not support the claim. When there is a clear ability to solve a problem with demonstrable business outcomes, you will stay on track.
What’s one way that GR8 People leveraged AI for their customers?
Every day the demand on recruiters intensifies. our AI solution to accelerate sourcing speed and accuracy, is the result of a team of GR8 People builders continuously working closely with our customers to evolve our platform in response to their biggest hiring challenges.
Our customers care about speed, quality, DEI, compliance, and the candidate experience. With GR8 People, they also don’t need to layer yet another expensive point solution over their core talent platform. Additionally, that eliminates challenges with data integrity and integration maintenance, as well as ongoing costs associated with per-match fee structures or profile limitations.
Leveraging the best available matching technologies, Source begins its work as soon as a recruiter opens a job, immediately identifying and automatically engaging the best talent for a role. We also made sure that recruiters can “see” exactly how matches are made—there’s a high level of transparency and control over the matching technology to actually refine search criteria for the best possible unbiased results.
The net outcome is a constantly filled pipeline of highly matched talent for every job, and even better–smart, automated engagement of high potential job seekers that results in faster hires.