What is Algorithm Aversion?
People tend to trust human judgment over algorithms. Even when an algorithm consistently beats human judgment, people prefer to go with their gut.
In 2014 Dietvorst, Simmons and Massey published an interesting article with the title: “Algorithm aversion: people erroneously avoid algorithms after seeing them err“. Through several studies they show that people are not as tolerant for algorithms as they are for people. If an algorithm makes a mistake, the confidence in the algorithm quickly goes down. You can see the mechanism at work when there is an accident with a Tesla; than the newspaper report how unreliable the automatic steering software is.
Dietvorst et. al. state: “The aversion to algorithms is costly (-) for society at large. Many decisions require a forecast, and algorithms are almost always better forecasters than humans”. Algorithm aversion is “enormously problematic, as it is a barrier to adopting superior approaches to a wide range of important tasks”.
Why is Algorithm Aversion relevant for HR?
In the figure below you can find examples of artificial intelligence in consumer and business (source: “How Artificial Intelligence is disrupting your organisation“).
Many of these examples touch the HR domain. Some examples of questions that can be asked:
- Will the employees trust the answers of the chatbot in the HR Service Centre?
- Will the management team buy the outcomes of the sophisticated people analytics as presented by HR?
- Will the supervisor have some patience with the robot who must learn new skills?
- Will the guard act on the security warning as presented by the intelligent security system?
How to deal with Algorithm Aversion?
This is a bias that will be difficult to overcome quickly. It might be wise to use the algorithms inconspicuously, so that the human factor cannot intervene.
When you sat next to someone in a Tesla you probably know the experience: if the ‘driver’ does not keep his or her hands on the steering wheel, you feel very uncomfortable. If you cannot see the driver, you care less (like in a metro without a driver).
Algorithms are increasingly used in recruitment. If the final human selectors only get to see candidates that are suitable for the job anyway, the human decision will have less impact (only on the selected candidate).
Dietvorst et. al published a follow up article in 2016: “Overcoming algorithm aversion: people will use imperfect algorithms if they can (even slightly) modify them“. They conclude that “one can reduce algorithm aversion by giving people some control -even a slight amount- over an imperfect algorithm’s forecast“.
Most likely the effect of algorithm aversion will diminish over time. The question is how steep the learning curve of people will be. In the meantime, it is wise to take algorithm aversion into account.
Illustration: Studio Fee Overbeeke