How can we know if AI negatively or positively affects enterprises, employees, and job candidates? Jolen Anderson, global head of human resources at BNY Mellon, says that “sometimes organizations decide not to leverage AI rather than investigate and determine if AI can be used for a purpose that drives positive outcomes.” This article will share some of those “positive outcomes” and why now is the right time for AI in the talent field.
MASSIVE DATA NECESSARY
The more advanced the AI, the more that it can do. The most advanced class of artificial intelligence is called neural networks or deep learning.
One of the primary challenges of powering deep-learning AI is the massive amount of data required. When analyzing talent, this requires billions of data points about people, career trajectories, capabilities, and experiences.
Many companies have tried to claim the mantle of ‘AI.’ Using only their own historical, limited pool of data results in a biased output. In order to ensure that bias is filtered out, it is necessary to analyze billions of data points across geographies, industries, and companies. For example, analyzing “successful” profiles in one company’s workforce as a model for future hires could result in bias that reflect historical biases, favoring limited demographic groups. Deep learning AI with equal opportunity algorithms can filter out any such bias.
A NEW WAY OF THINKING ABOUT SKILLS
With nearly 2 million unique skills globally, it is impossible for a human to understand all of them, let alone the correlations between them, and the trends of new and emerging skills.
And there has historically been little to no use of “adjacent skills.” These are skills that offer indicators of success in a different skill; a person good at A often excels at B.
Now, with artificial intelligence, we can infer that someone who excels in calculus likely also excels at algebra, but the reverse is not necessarily true. We can be confident that an experienced enterprise sales representative could quickly learn enterprise partnerships.
DEEPER AND FAIRER
Michael Ross is Visa’s former CHRO. He says that “companies want to know that decisions are being made with the best, most relevant data possible. And candidates want to know they are being represented for their skills, capabilities, experiences, and potential.”
But under the status quo, job seekers often find themselves wondering how their capabilities and potential fit into open roles. Should they apply for an Associate or a Senior Associate role? With AI, candidates can see how their capabilities match up to the needs of the job.
“When done right, AI safely makes data more complete and transparent,” says Dr. Huggy Rao, the Atholl McBean professor of organizational behavior and human resources at the Stanford Graduate School of Business.
Indeed, AI helps end the anachronistic overreliance on resumes, job descriptions, and interviews. And, the over-emphasis on “who you know.” Instead, it allows talent leaders to make much more intelligent employment-related decisions based on who’s capable of what.
THE ELEMENT OF TIME
AI can capture the dimension of time. A human recruiter, seeing General Electric on a candidate’s resume, may think “GE is a good company to have worked for” without any further understanding. AI can know that a job in a certain team or department of GE during a certain time period was different than one on a different team or at a different time.
Dr. Rao says this time element is not examined enough: “Everyone is going to succeed. And everyone, at some point, is going to fail. But how do you respond? You did a certain thing when the economy was great—how can we know if that was you, or the tailwinds of the macroeconomic forces?”
AI is now being used to increase internal mobility, avoid layoffs, better manage contingent work, improve non-traditional placement, and analyze skill gaps in a company’s workforce.
This article is excerpted from “AI for Good: A Better, More Inclusive Future of Work”. To continue reading more from that report, go here.