Wednesday, June 12, 2024

AI: Large Language Models Beginning To Understand And Deploy Deception Strategies

From the journal Proceedings of the National Academy of Sciences of the United States of America, June 4:

Deception abilities emerged in large language models

Significance
This study unravels a concerning capability in Large Language Models (LLMs): the ability to understand and induce deception strategies. As LLMs like GPT-4 intertwine with human communication, aligning them with human values becomes paramount. The paper demonstrates LLMs’ potential to create false beliefs in other agents within deception scenarios, highlighting a critical need for ethical considerations in the ongoing development and deployment of such advanced AI systems.

Abstract
Large language models (LLMs) are currently at the forefront of intertwining AI systems with human communication and everyday life. Thus, aligning them with human values is of great importance. However, given the steady increase in reasoning abilities, future LLMs are under suspicion of becoming able to deceive human operators and utilizing this ability to bypass monitoring efforts. As a prerequisite to this, LLMs need to possess a conceptual understanding of deception strategies. This study reveals that such strategies emerged in state-of-the-art LLMs, but were nonexistent in earlier LLMs. We conduct a series of experiments showing that state-of-the-art LLMs are able to understand and induce false beliefs in other agents, that their performance in complex deception scenarios can be amplified utilizing chain-of-thought reasoning, and that eliciting Machiavellianism in LLMs can trigger misaligned deceptive behavior. GPT-4, for instance, exhibits deceptive behavior in simple test scenarios 99.16% of the time (P < 0.001). In complex second-order deception test scenarios where the aim is to mislead someone who expects to be deceived, GPT-4 resorts to deceptive behavior 71.46% of the time (P < 0.001) when augmented with chain-of-thought reasoning. In sum, revealing hitherto unknown machine behavior in LLMs, our study contributes to the nascent field of machine psychology....

....MUCH MORE