AUTHOR=Edwards Darren J. 

TITLE=A functional contextual, observer-centric, quantum mechanical, and neuro-symbolic approach to solving the alignment problem of artificial general intelligence: safe AI through intersecting computational psychological neuroscience and LLM architecture for emergent theory of mind

JOURNAL=Frontiers in Computational Neuroscience

VOLUME=Volume 18 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/computational-neuroscience/articles/10.3389/fncom.2024.1395901

DOI=10.3389/fncom.2024.1395901

ISSN=1662-5188

ABSTRACT=Neuro-symbolic natural language processing density clusters, a conscious quantum level defined by QBism, and real-world applied level (human user feedback). It is argued that through this approach, AI could be conscious, form deictic perspective-taking frame abilities, thus forming human-level self-awareness, empathy, and compassion towards others. Crucially, this consciousness hypothesis is directly testable at approximately a 5-sigma significance (with a 1 in 3.5 million that the probability that any AIconscious observations that are identified are due to chance factors) via double-slit intenttype experimentation and visualizing procedures for derived perspective-taking relational frames. This could ultimately lead to a solution to the alignment problem and emergent theory of mind (ToM) within AI.