
The advent of generative artificial intelligence (GenAI) has seen much debate in academic and professional circles of late, as well as significant hype in public media and heated discussion on social media. Much of this attention has been negative and focused on its potential risks. In this paper, Hagendorff undertakes a review of references to GenAI in the published literature, and attempts to provide a balanced analysis of the issues.
Hagendorff begins with some background, including an explanation of the scope, an introduction to the ethical issues surrounding GenAI, and the potential issues associated with misaligned or poorly developed AI systems.
The methods used for the selection and assessment of papers in the review are presented and discussed, along with how papers were screened for inclusion. The results are then presented, covering many aspects of GenAI: bias, privacy, robustness, transparency, copyright, governance, and regulation.
The paper concludes with a discussion of the results, including conclusions and research limitations. A perceived imbalance in the literature and an exploration of unsubstantiated scenarios are covered. The paper includes thorough references and a uniform resource locator (URL) to supplementary material, and aspects of the review are clarified in several appendices.
Hagendorff has prepared a reasonably balanced overview of contemporary literature and the various ethical debates currently surrounding the use of GenAI, which should prove useful for academics, professionals, and policymakers.