Does It Make Sense to Speak of Introspection in Large Language Models?
This addresses a philosophical and interpretative problem for researchers and practitioners in AI and cognitive science, but it is incremental as it builds on existing debates about LLM capabilities.
The paper tackles the problem of whether the concept of introspection can be meaningfully applied to large language models (LLMs) by critiquing two examples of self-reports, concluding that one is not valid while the other represents a minimal form of introspection without consciousness.
Large language models (LLMs) exhibit compelling linguistic behaviour, and sometimes offer self-reports, that is to say statements about their own nature, inner workings, or behaviour. In humans, such reports are often attributed to a faculty of introspection and are typically linked to consciousness. This raises the question of how to interpret self-reports produced by LLMs, given their increasing linguistic fluency and cognitive capabilities. To what extent (if any) can the concept of introspection be meaningfully applied to LLMs? Here, we present and critique two examples of apparent introspective self-report from LLMs. In the first example, an LLM attempts to describe the process behind its own "creative" writing, and we argue this is not a valid example of introspection. In the second example, an LLM correctly infers the value of its own temperature parameter, and we argue that this can be legitimately considered a minimal example of introspection, albeit one that is (presumably) not accompanied by conscious experience.