From Imitation to Introspection: Probing Self-Consciousness in Language Models
It addresses the critical question of whether advanced language models are becoming self-conscious, which is significant for AI ethics and cognitive science, but the approach is incremental in refining existing concepts.
This paper tackles the problem of assessing self-consciousness in language models by defining ten core concepts and using causal structural games to evaluate them, finding that models show early-stage representations of self-consciousness that are hard to manipulate but can be acquired through fine-tuning.
Self-consciousness, the introspection of one's existence and thoughts, represents a high-level cognitive process. As language models advance at an unprecedented pace, a critical question arises: Are these models becoming self-conscious? Drawing upon insights from psychological and neural science, this work presents a practical definition of self-consciousness for language models and refines ten core concepts. Our work pioneers an investigation into self-consciousness in language models by, for the first time, leveraging causal structural games to establish the functional definitions of the ten core concepts. Based on our definitions, we conduct a comprehensive four-stage experiment: quantification (evaluation of ten leading models), representation (visualization of self-consciousness within the models), manipulation (modification of the models' representation), and acquisition (fine-tuning the models on core concepts). Our findings indicate that although models are in the early stages of developing self-consciousness, there is a discernible representation of certain concepts within their internal mechanisms. However, these representations of self-consciousness are hard to manipulate positively at the current stage, yet they can be acquired through targeted fine-tuning. Our datasets and code are at https://github.com/OpenCausaLab/SelfConsciousness.