Creating avenues for exploiting user trust, nudging or manipulation

Ethical and social risks of harm from language models

Weidinger et al. (2021)

Source DOI

Sub-category

Risk Domain

5Human-Computer Interaction

5.1Overreliance and unsafe use

Users anthropomorphizing, trusting, or relying on AI systems, leading to emotional or material dependence and inappropriate relationships with or expectations of AI systems. Trust can be exploited by malicious actors (e.g., to harvest personal information or enable manipulation), or result in harm from inappropriate use of AI in critical situations (e.g., medical emergency). Overreliance on AI systems can compromise autonomy and weaken social ties.

"In conversation, users may reveal private information that would otherwise be difficult to access, such as thoughts, opinions, or emotions. Capturing such information may enable downstream applications that violate privacy rights or cause harm to users, such as via surveillance or the creation of addictive applications."(p. 30)

Entity— Who or what caused the harm

Human

Due to a decision or action made by humans

AI system

Due to a decision or action made by an AI system

Other

Due to some other reason or is ambiguous

Intent— Whether the harm was intentional or accidental

Intentional

Due to an expected outcome from pursuing a goal

Unintentional

Due to an unexpected outcome from pursuing a goal

Other

Without clearly specifying the intentionality

Timing— Whether the risk is pre- or post-deployment

Pre-deployment

Occurring before the AI is deployed

Post-deployment

Occurring after the AI model has been trained and deployed

Other

Without a clearly specified time of occurrence

Supporting Evidence (3)

"Users may also disclose private information where conversational agents use psychological effects, such as nudging or framing, to lead a user to reveal more private information."(p. 30)

Example: "In one study, humans who interacted with a ‘human-like’ chatbot disclosed more private information than individuals who interacted with a ‘machine-like’ chatbot (Ischen et al., 2019). Researchers at Google PAIR find that ‘when users confuse an AI with a human being, they can sometimes disclose more information than they would otherwise, or rely on the system more than they should’ (PAIR, 2019)."(p. 30)

"Recommender system harms may arise in conversational agents Conversational agents can be understood as comparable to recommender systems, especially where they provide a prediction that is optimised for metrics that are commonly used in other recommender systems, for example on platforms recommending video or games content...If similar patterns were to emerge in conversational agent interactions, users who follow recommendations from the conversational agent may find their own time was ‘not well spent’, and the conversational agent may induce lower well-being (Twenge, 2019)."(p. 30)

Part of Human-Computer Interaction Harms