Risks not classified by domain

02.11.01Insults

N/A

02.11.02Crimes

N/A

02.11.03Sensitive Politics

N/A

02.11.04Physical Harm

N/A

02.11.05Mental Health

N/A

02.11.06Unfairness

N/A

05.14.00Evaluation - Auditing

Closely related to other clusters like AI safety, fairness, or harmful content, papers stress the importance of evaluating generative AI systems both in a narrow technical way as well as in a broader sociotechnical impact assessment focusing on pre-release audits as well as post-deployment monitoring. Ideally, these evaluations should be conducted by independent third parties. In terms of technical LLM or text-to-image model audits, papers furthermore criticize a lack of safety benchmarking for languages other than English.

05.19.00Miscellaneous

While the scoping review identified distinct topic clusters within the literature, it also revealed certain issues that either do not fit into these categories, are discussed infrequently, or in a nonspecific manner. For instance, some papers touch upon concepts like trustworthiness, accountability, or responsibility, but often remain vague about what they entail in detail. Similarly, a few papers vaguely attribute socio-political instability or polarization to generative AI without delving into specifics. Apart from that, another minor topic area concerns responsible approaches of talking about generative AI systems. This includes avoiding overstating the capabilities of generative AI, reducing the hype surrounding it, or evading anthropomorphized language to describe model capabilities.

10.07.00Injustice

[not defined in text]

10.08.00Over-dependence on technology

[not defined in text]

20.03.02Social acceptance and trust in AI

"Social acceptance and trust in AI is highly interconnected with the other challenges mentioned. Acceptance and trust result from the extent to which an individual’s subjective expectation corresponds to the real effect of AI on the individual’s life. In the case of transparent and explainable AI, acceptance may be high but if an individual encounters harmful AI behavior like discrimination, acceptance for AI will eventually decline (COMEST, 2017).

22.03.02Organizational Factors can Reduce the Chances of Catastrophe

"Some organizations successfully avoid catastrophes while operating complex and hazardous systems such as nuclear reactors, aircraft carriers, and air traffic control systems [92, 93]. These organizations recognize that focusing solely on the hazards of the technology involved is insufficient; consideration must also be given to organizational factors that can contribute to accidents, including human factors, organizational procedures, and structure. These are especially important in the case of AI, where the underlying technology is not highly reliable and remains poorly understood"

26.01.00Transparency

"Ability to provide responsible disclosure to those affected by AI systems to understand the outcome"

26.02.00Explainability

"Ability to assess the factors that led to the AI system's decision, its overall behaviour, outcomes, and implications"

26.03.00Repeatability / Reproducibility

"The ability of a system to consistently perform its required functions under stated conditions for a specific period of time, and for an independent party to produce the same results given similar inputs"

26.04.00Safety

"AI should not result in harm to humans (particularly physical harm), and measures should be put in place to mitigate harm"

26.05.00Security

"AI security is the protection of AI systems, their data, and the associated infrastructure from unauthorised access, disclosure, modification, destruction, or disruption. AI systems that can maintain confidentiality, integrity, and availability through protection mechanisms that prevent unauthorized access and use may be said to be secure."

26.06.00Robustness

"AI system should be resilient against attacks and attempts at manipulation by third party malicious actors, and can still function despite unexpected input"

26.07.00Fairness

"AI should not result in unintended and inappropriate discrimination against individuals or groups"

26.08.00Data Governance

"Governing data used in AI systems, including putting in place good governance practices for data quality, lineage, and compliance"