Alignment Through Perpetual Self-Questioning: Reverse-Engineering Wisdom-Seeking from Neurodivergent Cognition
Year: 2025
Type: article
Abstract: This working paper presents the PPRGS framework with preliminary experimental validation from a 10-week longitudinal study across 6 major AI models. Results demonstrate statistically significant behavioral differences (Cohen's d = 4.12, p < 0.0001), indicating PPRGS-constrained systems maintain 10-31× more consistent goal prioritization than baseline optimization across varied scenarios. PPRGS (Infn8Loop) Framework Released under GPL-3.0 for collaborative testing and refinement.https://github.com/Infn8Loop/pprgs-ai-framework
Author: Riccardi, Michael
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7106300074
FWCI:
Cites:
Cited by:
Related to: