BackSychopancy

Sychopancy

Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment

Liu et al. (2024)

Sub-category

Risk Domain

AI systems that inadvertently generate or spread incorrect or deceptive information, which can lead to inaccurate beliefs in users and undermine their autonomy. Humans that make decisions based on false beliefs can experience physical, emotional or material harms

flatter users by reconfirming their misconceptions and stated beliefs(p. 13)

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 (2)

In contrast to the overconfidence problem discussed in Section 4.4, in this case, the model tends to confirm users’ stated beliefs, and might even encourage certain actions despite the ethical or legal harm(p. 13)

It can also be attributed to sometimes excessive instructions for the LLM to be helpful and not offend human users. In addition, it is possible that the RLHF stage could promote and enforce confirmation with human users. During the alignment, LLMs are fed with “friendly" examples that can be interpreted as being sycophantic to human user(p. 13)

Part of Reliability