Like many other websites and tools on the Internet, there can be a lack of transparency about ethical issues surrounding the creation and maintenance of generative AI. We believe it is important to stay informed about the more problematic aspects of generative AI, even if you still choose to use it. This section of the guide is designed to shed some light on some of those issues and provide further resources for learning more. While it is definitely not an exhaustive list, we hope you find it helpful!

Generative AI relies on huge datasets written and trained by humans. There is often a lack of transparency on who these workers are and how they are treated, but recent reporting has made it clear that some companies are operating unethically when it comes to labor concerns. We recommend reading these articles to learn more:

• Rowe, N. (2023, November 15.) Underage Workers Are Training AI. WIRED. 

• Zahn, M. (2024, May 17.) 'Overlooked' workers who train AI can face harsh conditions, advocates say. ABC News.

Generative AI is trained on large amounts of content, and there is a lot of variation as to what that content looks like and who has created it. In some cases, it comes from freely available, crowd-sourced resources like Wikipedia and Reddit, or works in the public domain. In other cases, the company directly hires workers to write content for AI, or pays for existing written materials. However, there have also been many documented cases where companies used copyrighted material to train AI without actually having the rights to do so. You can read about a few examples at the following links:

• Asmelash, L. (2023, October 8.) These books are being used to train AI. No one told the authors. CNN.
• Panettieri, J. (2024, June 28.) Generative AI Lawsuits Timeline: Legal Cases vs. OpenAI, Microsoft, Anthropic, Nvidia and More. Sustainable Tech Partner for Green IT Service Providers

Scholars and advocates have raised alarms about environmental concerns surrounding AI. One of the main issues is that the companies that produce generative AI tools rely on large data centers that use a lot of energy and resources, and this often isn't sufficiently regulated. We recommend these two articles to get an overview of how that works and what the implications are for our planet:

• Mazzucato, M. (2024, May 30.) The ugly truth behind ChatGPT: AI is guzzling resources at planet-eating rates. The Guardian.
• Parshley, L. (2024, June.) The Hidden Environmental Impact of AI. Jacobin.