Generative ai could have a notable environmental footprint, and this story looks at what we stand to gain in return.
Generative ai is getting a lot of attention these days. ChatGPT reportedly has hundreds of millions of users, and similar functionality is supposedly being integrated into a multitude of digital products ranging from Microsoft Word and Teams to search engines.
The environmental footprint of generative ai could be noticeable if billions of people began using it extensively on a daily basis (1).
But will the value we get from this technology outweigh the potential environmental costs?
That is the question on which I will try to shed some light in this article, outlining some perspectives on what we can gain.
First, I will provide some insights into the potential productivity gains from generative ai.
Next, I will discuss whether generative ai will be a net positive or net negative technology.
Next, I will analyze the extent to which generative ai can reduce inequality.
Finally, I will provide my perspective on whether we should expect generative ai to accelerate the green transition.
Scope
Please note that what follows refers only Generative ai. Generative ai differs from other types of ai.
Also note that I focus only on generative ai that generates text (including code) and ignore image and sound generation technology.
For a simple working definition of generative ai, see (2).
Let’s consider the extent to which generative ai could generate productivity gains. A study on the effect of ChatGPT on productivity in various writing and analysis tasks found that…
