MIT AI Risk Repository · Risk Sub-Category · 33.02.01

Hallucination

Category: Technology concerns

Open the MIT Risk Navigator All entries from Nah2023

Description

"Hallucination is a widely recognized limitation of generative AI and it can include textual, auditory, visual or other types of hallucination (Alkaissi & McFarlane, 2023). Hallucination refers to the phenomenon in which the contents generated are nonsensical or unfaithful to the given source input (Ji et al., 2023). Azamfirei et al. (2023) indicated that "fabricating information" or fabrication is a better term to describe the hallucination phenomenon. Generative AI can generate seemingly correct responses yet make no sense. Misinformation is an outcome of hallucination. Generative AI models

From Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration (Nah2023), as extracted by the MIT AI Risk Repository (CC BY 4.0).

Classification

Domain: 3. Misinformation
Subdomain: 3.1 False or misleading information
Causal entity: AI
Intent: Unintentional
Timing: Post-deployment

Subdomain definition: 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

Real-world incidents in this subdomain

Browse all incidents in this subdomain

How other frameworks describe this risk

Pursuing Consistent Context
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Knowledge Gaps
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Hallucinations
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Faithfulness Errors
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Defective Decoding Process
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Untruthful Content
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Factuality Errors
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Noisy Training Data
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

Other entries from Nah2023

Ethical Concerns Risk Category
Harmful or inappropriate content Risk Sub-Category · 1.2
Bias Risk Sub-Category · 1.1
Over-reliance Risk Sub-Category · 5.2
Misuse Risk Sub-Category · 4.3
Privacy and security Risk Sub-Category · 2.1
Digital divide Risk Sub-Category · 6.1
Technology concerns Risk Category · 7.3
Quality of training data Risk Sub-Category · 7.3
Explainability Risk Sub-Category · 7.4
Authenticity Risk Sub-Category · 6.3
Prompt engineering Risk Sub-Category · 7.4