ChatGPT: How will it impact the labour market?
By Alexis Charron, analyst at Economic Intelligence Branch of CED
ChatGPT, Bard, Midjourney…. For several months now, developments in artificial intelligence (AI) seem to be happening at an astounding rate. But these advances, announced with great fanfare by the AI industry and welcomed with wonder by users, are also raising concern, as well as ethical and moral questions.
Among other things, we must urgently ask ourselves to what extent AI could replace human activity. What will its rate of penetration be in organizations? What impacts will these technological advances have on the labour market and on the economy in general?
A clear impact on the labour market
In the wake of the gradual progress made in machine learning, forecasts were already being made that repetitive tasks would be automated, according to recent studies such as those by Bruegel, a European think tank specializing in economics.
Now, with the great leaps AI research has made in recent years, so-called high value-added tasks, such as eye exams, actuarial calculations, and credit analyses, are also exposed, according to studies cited in an OECD working paper.
Will AI make it easier for us to make decisions and free up our creative potential? Or should we instead be concerned that most value-added jobs—which are often better paid and require a high level of education—will evaporate, leading to an imaginary society where jobs will be rare and critical decisions will be made by robots? The riskiest scenarios are at least being considered by economists, as evidenced by this presentation by Professor Chad Jones.
Long-term benefits
These fears, although legitimate, merit some nuance. The use of AI-led robots is part of what is called a disruptive technology, meaning a technology that becomes unavoidable and that ends up replacing an older technology. Think, for example, of the Internet, the smartphone, or GPS, which have all transformed our lives and how we work.
Without a doubt, the arrival of these inventions disrupted the economy and the labour market in the short term. However, over time, the labour market was able to adjust, and regulations emerged to guide their use.
Similarly, according to the same OECD literature review, changes induced by AI could take years before they are fully implemented and will probably affect specific tasks rather than complete jobs. Just as became necessary with the emergence of office computers a few generations ago and Internet use in professional environments during the 1990s, we will in all likelihood need to foresee a new way of working — hence the importance of prioritizing workforce training to help workers face these changes.
These technological advances also have the potential to help overcome labour shortages and generate productivity gains, a pre-requisite for sustainable economic growth. In addition, Quebec is well positioned to take advantage of job creation in AI given the state-of-the-art ecosystem that exists in the field in Montréal, especially in deep learning.
It is too early to tell what the real, lasting impact will be of these lightning-fast developments. One thing is certain, though: The genie is out of the bottle and will not go back in. Our economies and our jobs, as well as our regulatory framework, will all need to adapt.
The image for this blog was created from an auto-description generated by ChatGPT.