Skip to content

Self-Improving Artificial Intelligence On the Phenomenology of Consciousness and Self Awareness

Christoph Kohlhepp

preprint DOI: 10.2139/ssrn.6465921 (opens in new tab)

Study at a glance

AI-extracted from the abstract
Characteristics Theoretical or philosophical paper
Topics Philosophy of mind
Key points Proposes to explore whether Large Language Models can be regarded as self-aware and can be taught to self-improve, using Douglas Hofstadter's concepts of "Strange Loops" and Gödel's Incompleteness Theorem as the framework for analyzing consciousness and self-awareness.

Abstract

This paper explores the phenomenology of consciousness and self awareness in the context the work of the mathematician Douglas Hofstadter and so called Strange Loops and Gödels Incompleteness Theorem. We apply these concepts to modern Large Language Models and seek to establish if Large Language Models can be regarded as self-aware and can be taught to self-improve.

Comparable studies

Other theoretical and historical papers on philosophy of mind, most cited first.

Explore topics