This Coalition of Companies Promises to End Algorithmic Bias in Hiring

The reality we all face today is that just about every company is now a data and technology company. Even if an enterprise’s business model is focused on making things, technology is largely driving the process. Of course, that means many companies’ operations are running on bits and bytes. That isn’t a bad thing, necessarily, but one problem is that as technologies such as artificial intelligence have roles in helping companies function day to day, there are always some risks. Among them are the impacts that algorithmic bias can have on a company’s hiring process — which are not great optics at a time when companies are striving to boost (or ahem, start) their diversity, inclusion and equity (DEI) efforts.

To that end, a coalition of U.S. companies has pledged to stop algorithmic bias from having any effects on their hiring practices.

The Data and Trust Alliance is comprised of more than a dozen companies that have committed to taking steps to ensure that algorithmic bias is removed from the ways in which they hire people. The corporations that are a part of this effort include American Express, Deloitte, General Motors, IBM, Mastercard, Nielsen, Nike and Under Armour.

As part of the Alliance, participating companies acknowledge the reality: Algorithms, artificial intelligence and data have huge influence in day-to-day and long-term business decisions. Nevertheless, just as how more companies are becoming aware of how their operations affect the planet, there is also that very human element — including the truth that many people have been excluded from various opportunities because such 21st-century technologies have screened them out (and continue to do so).

Announced only yesterday, these brands have so far been vague on how they will manage algorithmic bias so that it is eliminated from any hiring decisions. The three pillars of their commitment include sharing best practices, i.e., how they are getting this done; new tools to harness algorithms, artificial intelligence and data for the better; and to be clear, this is a working group, not a “think tank” or lobbying group.

To start, the Alliance has released a set of guidelines framed by 55 questions spread across 13 categories, the goals of which include preparing companies so that they can better vet vendors on various criteria, including transparency, ethics and testing for bias.

And while left unsaid, there is the wider goal to secure these companies’ brand reputations. No one wants to be in the middle of a social or traditional media firestorm with reports of certain demographics being shafted from new jobs or promotions yet again: So to prevent those fires, get rid of any kindling, as in such data or algorithms that could end up causing bias before a human can even see a person’s human resources (HR) information or a resume.

“As a leading data and analytics company, we know that equitable and unbiased data is essential to trust and truth,” said David Kenny, CEO of Nielsen. “The Alliance’s anti-bias criteria for HR algorithms are essential to trust and truth when hiring and developing a winning team.”

Image credit: Van Tay Media via Unsplash