Detail

Opinions - 03.02.2020 - 00:00 

Johanna Gollnhofer on the new decade: "Alexa teaches students"

The start of the new decade is accompanied by an increase in predictions. But do they make sense and how reliable are they? HSG professor Johanna Gollnhofer traces these questions in an opinion piece. Her prediction for the coming years: voice assistants such as Alexa will teach students.
Source: HSG Newsroom

3 February 2020. The first iPhone appeared in 2007. Sneered at by rivals at the time, it has now completed its triumphal procession into our everyday lives. My screen time tracking speaks volumes: on average each day, barely 30 minutes go by without me picking up my iPhone at least once. In contrast, other technologies which have been pitched as game-changers barely get used. Segways, for example, are mostly used these days as a way for tourists to get around. The inventor's vision for them, however, was that they would radically change the future of city centres by replacing cars. 

Candidate for trends

What will life be like for us in the future? Will there be one or more new technologies which will radically change our everyday lives? The list of technological candidates is long: speech technologies, robots, Google glasses, autonomous devices, smart products, autonomous cars, virtual reality glasses or even chips which are implanted under our skin? Or will e-scooters, which took over European cities in 2019, fulfil the Segway vision?

At the start of 2020, predictions amassed as to which technologies would gain the market share and shape our lives. For example, Google experts announced that, among others, voice assistants and visual searches will have an enduring influence on the consumer world and also entire industries. 

A sense of security

The point of predictions is easy to grasp: they give us a sense of security in an ever changing world – even if this sense is illusory. Even in our everyday lives, predictions and assumptions play an important role. Some of these are more accurate than others: The next train is running on time versus I will meet my dream man today. 

Predictions should always be treated with the utmost caution: in 1937, a large-scale study predicted that half of the American population would be living in trailers in the next 15 to 20 years. The basis for this prediction appeared to be plausible: the demand for trailers at this time was greater than the supply and the assumption made by the experts was that this trend would continue. 

Exciting thought experiments

The "Advantage" of predictions is that they are usually not checked retrospectively. For example, at the end of 2020, who is going to check whether the predictions made by the Google experts actually came true?

That is why I can confidently voice my prediction for the coming years: voice assistants will teach students. In my opinion, it is firstly irrelevant whether this prediction will actually come true. What I find far more exciting are the thought experiments that allow it. 

Picture: Adobe Stock / metamorworks

north