It’s not often that we find artificial intelligence (AI) in the company of a U. S. Supreme Court Chief Justice, a prestigious all-girls high school and a good joke – which itself sounds like a classic setup to a joke!  But we did.

At his daughter’s high school graduation, Chief Justice John Roberts recently opined on a number of thoughtful topics – including the risk of getting caught in an echo chamber of our personalized feeds and messaging, telling us what we want to hear and blocking out the formation of new ideas.

As proof that “AI is everywhere,” Roberts hinted at the possibility that AI and big data can and has influenced elections. Moreover, this potent combo reinforces what we want to hear by targeting us with what savvy marketers and political consultants want us to hear, leading to an infinite loop (my words).  “The result,” he said, “can be a narrowing and over-simplification that is contrary to individuality and creativity.” 

Heeding the Warning

In addition to the political implications, there’s also a business-practical side to Roberts’ warning.  He’s spot-on with a key observation: One of the potential unintended side effects that we should watch for in a big data- and AI-driven world is precisely this “narrowing and over-simplification.”  If AI decides what goes on the retail shelf, and the sales from the retail shelf inform AI what to keep and what to drop, the universe of options narrows down to the familiar and repeated. 

What if, in 1975, AI analyzed market research and found that no one wanted a computer in their home? And say in 1980 that AI listened to automobile purchasers and decided that consumers weren’t asking for a minivan to replace their station wagon?  But when Chrysler began selling minivans in 1983, it brought in enough revenue to save the company. 

Only those with the vision to break the infinite loop of quantitative market feedback can create new markets with new ideas.  An auto industry quote attributed (incorrectly) to Henry Ford is too good to pass up: “If you asked people what they wanted, they would have said a faster horse.” Despite its flawed attribution, the quote hits a key point: It takes the creative human spirit, coupled with careful observation and understanding of human nature, to create new ideas and ensure that we don’t operate in an echo chamber of tailored stories feeding our tailored prejudices – be they politics, lifestyle choices or retail decisions. 

Breaking Free of the Infinite Loop

One remedy is to start from the bottom-up and examine basic human needs that are critical to sustaining a business.  For example, a bank should ask, “What do people think about money and financial security?” rather than, “What do they want in a banking app?” A healthcare insurance carrier should inquire, “What do people think about their health planning?” rather than, “What do they want in a claims management screen?” We’ve found that assembling the “thick data” (gained through anthropological study) that surrounds our lives, actions and motivations is a necessary complement to the big data that provides dashboard views of past behavior.

In his commencement address, Roberts described another remedy for breaking the trend of narrowing thought: taking personal quiet time to ponder your own thoughts rather than simply acquiring more information. This personal entreaty is probably just what the graduating class needed to hear – but for those of us in business, the advice must go deeper.

Organizations can start by assembling a team of talented people – ethicists, sociologists, human scientists, marketers, consumers and teachers – to assess every aspect of an AI- or big data-driven system, from algorithm creation through training.  An august group like this, guided by strategic consultants, can conduct a reality check and encourage brainstorming in a way that will shape the ethical use of AI and analytics for years to come and countermand infinite-loop thinking.

Businesses that have used this approach have reaped the rewards:

  • A major toy manufacturer with whom we worked realized that playtime was socially connected and that it reflected on personal skill and pride. It discovered its market was not price-sensitive and didn’t require a race to the bottom in quality and cost.
  • A credit card issuer with whom we partnered came to understand that small businesses considered their choice of credit supplier to be part of the fabric of their business success. Rather than being motivated solely by fees and features, they were also compelled by trust and experience.

Oh, and that joke I promised?  Upon learning he had only 10 minutes on the program, Roberts claimed to have asked the school’s head how he could possibly communicate all the wisdom he’d acquired over the years in that short of a time. To which, he said, she responded,  “Speak very slowly.”

Could AI could have written that line?  Not yet

James Jeude

James Jeude

James Jeude is Vice-President and Practice Leader of Cognizant Digital AI & Analytics Strategic Initiatives Group. His career has provided perspective for... Read more