Until just a few years ago, everyone was using books or encyclopaedia as their source of information. These tended to be well-researched, and, depending on the publishing house, were carefully checked for errors or omissions. These sources of information were also written by highly specialized experts. One could be virtually sure that the information provided was correct, as far as this is possible. While clearly some books are written with the intention of shifting public opinion, at least academic contributions or encyclopaedia were written to inform, and not to shift opinion.

Then the internet led to the development of collaboratively-written encyclopaedia such as Wikipedia, and suddenly everyone could add to the globally-accessible wealth of information. In the first years of Wikipedia everyone was questioning the content, but over the years, it seems that collaboratively-written sources such as Wikipedia have become a widely-accepted portal of information that is now viewed as the new standard in society. While this source of information is more easily accessible than books, it is also likely to lead to more misinformation (dissemination of false information by someone who believes it is true), disinformation (dissemination of false information by someone who knows it is false) or even malinformation (dissemination of false information by someone who wants to harm others).

The last couple of years has seen the rise of a new source of information: Artificial Intelligence. AIs such as ChatGPT have been in the news for some time now. They promise an enormous potential, and have been discussed extensively. I believe that, eventually, they could turn out to be amazing. However, right now, they are problematic, simply because nobody really knows where they obtain their information from,  how they collect it,  and how they transform it into the final text that we obtain. If one were to use this information without adequate check-up, then this could become a real issue. AIs such as ChatGPT obtain their information from unknown sources, and combine them in a – for most humans – unknown way. Depending on the AIs code, anything from misinformation over disinformation to malinformation could be possible. If we take the results of an AI query at face value, we may be prone to falling prey to any of these types of mistakes.

Let me give you an example. I take the most extreme case of assuming that I know very little about anything else, and the only knowledge that I really have is about myself. So I asked ChatGPT questions about myself and collected the answers below. I crossed out the parts of the answer that ChatGPT got wrong and then commented in brackets behind. This should give you an idea about the potential dangers of fully relying on an AI, where you simply do not know as to what were the sources of information and how the AI assembled these.

Who is Ingmar Schumacher?

Ingmar Schumacher is an academic economist who specializes in the fields of environmental and resource economics, and more broadly, the intersection of economics, ecology, and sustainability. He is currently a Professor of Environmental Economics at IPAG Business School in France and also holds the position of Visiting Professor at the University of Johannesburg in South Africa (I have no connection to the University of Johannesburg).

Schumacher’s research focuses on topics such as the economics of biodiversity, climate change economics, and sustainable development. He has published numerous articles in peer-reviewed journals and has also served as a consultant to various organizations including the United Nations Environment Programme (UNEP) and the European Commission (I have never worked as a consultant for UNEP or EC).

Could you give me more details about Ingmar Schumacher, please.

Sure, here are some additional details about Ingmar Schumacher:

  • Ingmar Schumacher was born in Germany and received his PhD in economics from the University of Luxembourg in 2010 (I received my PhD from Louvain-la-Neuve, Belgium, in 2007).
  • He has held academic positions at various universities around the world, including the University of Southern Denmark, the University of Leeds, and the University of Gothenburg (I never held a position at any of these).
  • In addition to his research and teaching activities, Schumacher has also been involved in various policy and outreach activities related to environmental economics and sustainability.
  • He is a member of several professional associations, including the European Association of Environmental and Resource Economists (EAERE) and the Society for Environmental Economics and Policy Studies (SEEPS) (nope not true).
  • Schumacher’s research has been cited extensively in academic literature, and he has also been interviewed by various media outlets, including the Financial Times and Bloomberg (I have written for the Financial Times but never for Bloomberg, and I have not been interviewed by the FT).
  • Some of his notable publications include “A model of knowledge management and the environment” (Journal of Economic Behavior & Organization, 2010), “Economic Growth and the Environment” (Journal of Economic Surveys, 2012), and “Economic growth and biodiversity: an overview of theoretical and empirical literature” (Environmental Science & Policy, 2018) (I published neither of these papers, which is a bit embarrassing as it would imply I don’t have notable publications).
  • Schumacher has received numerous awards and honors for his work, including the Best Paper Award from the European Association of Environmental and Resource Economists in 2015, and the Springer Thesis Award in 2011 for his doctoral thesis on the economics of renewable energy (I received neither of these, that would have been nice though). He is also a Fellow of the Royal Society of Arts (at least I am unaware of this but if asked I would definitely accept, so if anyone from the RSA reads this you know where to find me) and a member of the European Association of Environmental and Resource Economists.

What is Ingmar Schumacher’s position on climate change?

As an academic economist who specializes in environmental and resource economics, Ingmar Schumacher has written extensively on the topic of climate change and the economics of climate policy. His position on climate change can be summarized as follows:

  • Schumacher views climate change as a serious and urgent problem that requires immediate action to mitigate its impacts and reduce greenhouse gas emissions.
  • He believes that the most effective way to address climate change is through a combination of mitigation and adaptation strategies (I believe we should focus as much as possible on mitigation), and that these strategies should be based on a deep understanding of the complex interactions between the economy, the environment, and human behavior.
  • Schumacher argues that climate policies should be designed to be economically efficient and socially equitable, and that they should take into account the costs and benefits of different policy options, as well as the distributional impacts of these policies.
  • He advocates for the use of market-based instruments, such as carbon pricing and emissions trading, as a way to incentivize emissions reductions and drive innovation in low-carbon technologies.
  • Schumacher is also interested in the role of international cooperation in addressing climate change, and has written about the potential for global agreements and cooperation to facilitate emissions reductions and climate adaptation efforts.

Overall, Schumacher’s position on climate change is informed by his expertise in environmental and resource economics and his commitment to finding economically efficient and socially equitable solutions to the complex challenges posed by climate change.

So those were ChatGPT’s answers. While the discussion that ChatGPT provides on my position on climate change is not wrong (apart from the comment on adaptation), my main question is where it obtained this information? Yes, this vaguely reflects my position on climate change, but I have never written this out clearly. It is also so generic that it applies to most environmental economists who work on climate change.

Overall, most of the information that ChatGPT provided about me was wrong. And I should know, well, because I guess I am the best expert when it comes to myself. But what does this imply about the information that ChatGPT provides? How can we ever be sure that this is correct? Even vaguely?

Naturally, this leads to a discussion of whom to trust. Our preference would be towards an objective, unbiased expert. The closest to this should be academic writing. In the absence of this one may prefer collectively-written resources, as these at least contain information provided from a variety of people which hopefully leads to somewhat unbiased information.

While one may hope that AIs are built in such a way as not to disseminate disinformation or malinformation, the question remains how it establishes which information is correct in order not to misinform.