Professor Gulyas: How is AI going to change the labor market?
ChatGPT: Artificial intelligence will change the labor market in multifaceted ways. One aspect of this is that it will allow certain jobs to be automated. This could lead to job losses. The flip side of this coin is that new job opportunities will emerge, especially in the areas of developing AI systems, data analysis, and machine learning. A study by the World Economic Forum predicts that automation could displace about 85 million jobs by 2025 while simultaneously creating 97 million new jobs.
Source: “The Future of Jobs Report 2020” by the World Economic Forum.
Note by Gulyas: I can confirm that this is accurate. The argument is supported well by a trustworthy source.
Gulyas: What will the consequences of this be for individual workers who lose their jobs?
ChatGPT: Workers who lose their jobs due to processes being automated and handled by AI will face challenges. The consequences of job losses will vary from sector to sector and depend on people’s skills profiles. Here are some of the possible outcomes:
It is important to support political measures and educational initiatives that will help workers to adapt to changes in the work environment.
Source: “The Future of Jobs Report 2020” by the World Economic Forum and “Risks and Benefits of Artificial Intelligence” by Bessen, J. E. (2020) in NBER.
Note by Gulyas: ChatGPT uses the generic masculine form here (Arbeitnehmer are male workers) even though I expressly avoided wording my question in a way that included only masculine forms. But overall, this is a very good answer. I would add that it has been well documented by research that cushioning income losses after job losses is difficult. That is why ensuring that the employees of tomorrow have the right skills, skills that are difficult to replace with AI; will need to start early — while people are still at school. It also has been well documented that ChatGPT tends to “hallucinate” and invent sources. This answer shows this in action: I could not find this source and I can only assume that it does not exist, although it is true that the authors mentioned have written articles on this topic.
My conclusion: ChatGPT has done a good job here, although the strengths and weaknesses of Large Language Models (LLM) are clearly evident. ChatGPT supplies pretty plausible answers to the questions, and I can largely confirm their accuracy. At the same time, it is also true that numerous interviews like this one already exist — that explains why ChatGPT can tackle this task with such ease. ChatGPT predicts and generates its answers on the basis of existing texts, and that means that topics that have been covered extensively are the topics it deals best with. It seems safe to say that ChatGPT cannot completely replace humans, but that it can provide assistance with, say, drafting texts, although the results will certainly still need checking and revision by an expert. In future, however, basic interview questions can simply be addressed to ChatGPT instead of to me, and that will leave me more time for research.