Future Focus: The Trouble with Algorithms

By Jennifer Schramm Nov 1, 2013
Reuse Permissions

November Cover

An algorithm is a series of clear steps that result in a particular output. Increasingly, we are all surrounded by the output of countless algorithms. They shape our economic markets, our business processes and our understanding of data. Even HR—a people-oriented profession if ever there was one—is being influenced by the world of algorithms.

According to the Society for Human Resource Management’s Technology & HR Management Special Expertise Panel, one key trend to watch in the coming years is "the need for HR professionals to be able to harness and analyze big data to drive business performance results." As HR becomes more data-driven, a greater number of processes can be captured in a series of simple to complex algorithms that can be carried out with minimal human involvement. Thus, the march toward automation continues for many HR functions and tasks.

But some events have suggested that when a system becomes too dominated by algorithms, it can abruptly spiral out of control before humans can intervene to regain control. This appears to have been the case in recent months when a series of computer glitches brought operations in global financial markets to a grinding halt. In a study published online in September 2013 by Scientific Reports, University of Miami physics professor Neil Johnson and his co-authors examine this ultrafast machine ecology.

According to the authors, competition to develop superfast systems that can outpace competitors in response to market changes has been driving this phenomenon. The problem occurs when the algorithms operate so fast that humans are unable to participate in real time. In this scenario, conventional market theories no longer apply, resulting in an unpredictable and out-of-control sequence of events in which the algorithms begin to behave in a way that can be likened to a mob. The study authors believe that insights into these robotic cybermobs may have important implications outside of finance, such as in dealing with cyberattacks and cyberwarfare.

In the world of HR, these new insights have major implications. They could influence the way human resource information systems are designed to work together in the future. Many large organizations are shifting to an almost completely data-driven model for key HR functions such as recruiting. This isn’t necessarily a bad thing—many companies that have shifted to these kinds of models report impressive results—but safeguards need to be considered.

Increasingly, these systems are being connected to other data systems, potentially forming a kind of computational ecosystem not unlike those of the financial systems examined in the study. The lesson seems to be that the greater the reliance on such systems, the more care must be taken to make sure that they are robust and secure and that humans can intervene when needed.

As more jobs become automated or affected by advanced technologies, HR will need to take the lead in ensuring that highly skilled, knowledgeable and technologically literate employees are ready to react to and overcome potential problems that occur as a result of cyberattacks or high-speed cybermobs that veer out of control.

Jennifer Schramm is manager of the Workforce Trends and Forecasting program at SHRM.

Reuse Permissions


Join SHRM's exclusive peer-to-peer social network

Join Today

Job Finder

Find an HR Job Near You


Find the Right Vendor for Your HR Needs

SHRM’s HR Vendor Directory contains over 3,200 companies

Search & Connect