Doomsday Predictions Around ChatGPT Are Counter-Productive

The last few weeks have been abuzz with news and fears (well, largely fears) about the impact chatGPT and other generative technologies might have on the workplace. Goldman Sachs predicted 300 million jobs would be lost, while the likes of Steve Wozniak and Elon Musk asked for AI development to be paused (although pointedly not the development of autonomous driving).

Indeed, OpenAI chief Sam Altman recently declared that he was “a little bit scared”, with the sentiment shared by OpenAI’s chief scientist Ilya Sutskever, who recently said that “at some point it will be quite easy, if one wanted, to cause a great deal of harm”. Indeed, a report from the company itself suggested that “most” jobs will be at risk in some way due to their technology.

In a world where most tech companies tend to over-inflate the good their products can do, it’s incredible that the generative AI movement is using their hype to propagate apocalyptic doomsaying instead. This is an especially stark juxtaposition as the company has, since its founding in 2015, marked itself as a nonprofit team of humanitarian scientists working for the good of us all.

As Dan Greene, an assistant professor at the University of Maryland’s Information Studies College, recently explained, these predictions are often using the new tool to “tell its own fortune” and to “mark down a boundary for public debate”.

Hype vs reality

The warnings about the future of jobs are revisiting the flurry of concern that surrounded the publication of Frey and Osborne’s 2013 paper into how their generation of AI will impact employment. Not only have those breathless fears not come to pass but we’re actually in a period of historically low unemployment.

Indeed, a recent report from CompTIA shows that last year, the technology industry witnessed a noteworthy surge in employment across all states. The sector recorded a net increase of 3.2%, resulting in the addition of over 280,000 jobs nationwide. Such promising statistics attest to the continued expansion of the technology industry, positioning it as a vital contributor to the U.S. job market.

The report reveals that the technology sector in the United States employs a whopping 9.1 million individuals, spanning both technical and non-technical roles. According to the report, Texas proved to be the frontrunner in the creation of tech jobs, recording a commendable addition of approximately 45,000 positions in 2022. Following closely behind, California secured the second spot, having added 38,186 tech jobs, while Florida, New York, and Washington state followed with 22,029, 18,487, and 17,962 job gains, respectively. These results underscore the enduring strength of the technology industry across various regions in the U.S.

Feeling in control

One would not think the jobs market was in such rude health if you only listen to the hype surrounding chatGPT, however, and this distinction matters for our prospects in the future of work.

It is difficult to underestimate the importance of self-efficacy in personal development. This is backed up by research from the University of Basel, which found that the most important determining factor of success in both one’s education and career was one’s aspirations.

In a form of the Pygmalion Effect, they found that if people believed they could thrive then they were far more likely to do so. This is problematic, as research conducted in deprived communities elsewhere in England found a strong sense of pessimism around the future and the prospects for them.

Past research has also underscored self-efficacy as one of the most important constructs for motivation and behavioral change. It is one of the key elements of psychological capital, and it has been repeatedly shown to have positive effects on a number of personal and work-related outcomes.

Inflated concern

This matters, as a recent paper from Brigham Young University, found that we’re often prone to greatly exaggerate the risk of automation taking our jobs.

The study reveals that just 14% of workers say they’ve ever seen a human being replaced by a robot. There is, however, a sense among those for whom automation has befallen them to exaggerate the risk of this happening across the economy by around 300%.

While only around 14% of employees had suffered from automation themselves, those people were much more likely to believe others had also been replaced by machines. Indeed, these people thought nearly half of workers had been automated, which compares to a figure of just 29% among the entire sample.

“Overall, our perceptions of robots taking over is greatly exaggerated,” the researchers explain. “Those who hadn’t lost jobs overestimated by about double, and those who had lost jobs overestimated by about three times.”

Believing in the future

The doomsday narrative propagating at the moment also undermines the confidence we can have in the future, which is also crucial in underpinning our self-efficacy. Research from ESSEC highlights how interconnected our hopes for the future are and our ability to act in a way that benefits us.

The researchers found that there is a strong link between our aspirations for the future and our willingness and ability to take responsibility for our careers and to take preparatory measures to plan for the future, such as building skills, developing our professional networks, and so on.

In other words, when we have a negative perception of our future, we’re less likely to take the kind of steps required to actually prepare for whatever the future has in store for us. So even if the doomsday predictions are accurate, which I doubt, they are likely to result in people becoming less prepared and less able to adapt should they come to pass.

When it comes to feeding the hype cycle of new technologies, we should perhaps be mindful of crying wolf, especially if we truly wish to avoid the adaptation challenges posed in past periods of technological disruption.

Facebooktwitterredditpinterestlinkedinmail