AI deception: A survey of examples, risks, and potential solutions
By Peter S. Park, Simon Goldstein, Aidan O'Gara, Michael Chen, Dan Hendrycks. Patterns, 2023
369 citations recorded by the source corpus.
Summary This paper argues that a range of current AI systems have learned how to deceive humans. We define deception as the systematic inducement of false beliefs in the pursuit of some outcome other than the truth. We first survey empirical examples of AI deception, discussing both special-use AI systems (including Meta’s CICERO) and general-purpose AI systems (including large language models). Next, we detail several risks from AI deception, such as fraud, election tampering, and losing control of AI. Finally, we outline several potential solutions: first, regulatory frameworks should subject AI systems that are capable of deception to robust risk-assessment requirements; second, policymakers should implement bot-or-not laws; and finally, policymakers should prioritize the funding of relevant research, including tools to detect AI deception and to make AI systems less deceptive. Policymakers, researchers, and the broader public should work proactively to prevent AI deception from destabilizing the shared foundations of our society.
Subfields: Artificial Intelligence, Safety Research, Signal Processing
View Paper: https://www.semanticscholar.org/paper/afe455cfb03b9487c8f0b67993f42f834b633c23
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Percy Liang
- Dan Hendrycks
- Philip H. S. Torr
- Sergey Levine
- Quoc V. Le
- Bernt Schiele
- Luke S. Zettlemoyer
- Jianfeng Gao
In Safety Research
In Signal Processing
- Dan Hendrycks
- Mantas Mazeika
- Steven Basart
- Shai Shalev-Shwartz
- Philip H. S. Torr
- U. Tupakula
- Ponnurangam Kumaraguru
- Efstratios Gavves
Citations in the corpus
- The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning (2024, background): “…from LLMs. Recent work has highlighted safety concerns of language models, including generating falsehoods (Ji et al., 2023; Zhang et al., 2023), producing toxic content (Gehman et al., 2020; Deshpande et al., 2023; Pan et al., 2024), and deceiving humans (Park et al., 2023; Scheurer et al., 2023).”
- Managing extreme AI risks amid rapid progress (2023, background): “However, AI is making progress in critical domains such as hacking, social manipulation, and strategic planning 24,31 , and may soon pose unprecedented con-trol challenges.”
- Mechanistic Interpretability for AI Safety - A Review (2024, background): “This focus on cherry-picked results from toy models is a form of streetlight interpretability (Casper, 2023), examining AI systems under only ideal conditions of maximal interpretability.”
- The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets (2023, background): “Despite their impressive capabilities, large language models (LLMs) do not always output true text (Lin et al., 2022; Steinhardt, 2023; Park et al., 2023).”