This is an interesting article about real-time gas pricing using “AI Learning” algorithms.
What do you think about this?
Should companies be allowed to do this?
When the Machines Learn to Price-Gouge
The introduction of algorithmic pricing software by German gas stations may have led to collaboration rather than competition between the machines
Famously collaborative.
PHOTO: ACADEMY OF MOTION PICTURE ARTS AND SCIENCE/EUROPEAN PRESSPHOTO AGENCY
Handing over responsibility for pricing to an algorithm sounds like a way to avoid baser human commercial instincts. But what if the algorithm learns to rip you off anyway?
That is the suggestion posited by economists at Queen’s University in Ontario, whose study on artificial-intelligence software in retail gas markets indicates that algorithms learn to collaborate with each other, rather than to compete.
In 2017, many German gas stations introduced software that tweaked prices intraday, learning what customers would bear. Around that time, an interesting shift occurred in markets with just two gas stations—a duopoly, in economic terms.
In such markets where both stations appeared to adopt algorithmic software, as estimated by sudden changes in the size and rapidity of price changes, margins increased by an average of almost 30%. Without pricing software at both stations, margins were unchanged.
In 1776, Adam Smith warned that when people of the same industry meet, “The conversation ends in a conspiracy against the public, or in some contrivance to raise prices.” Little could he know that when the machines meet, they might end up doing the same.
Operations and Supply Chain Management 