Η διαλεκτική του πλατωνικού Σωκράτη μέσω του διαλογικού πράκτορα PlaSoC

Παπαδήμας, Μιχαήλ

The dialectic of the Platonic Socrates through the dialogical agent PlaSoC

Master Thesis

Author

Παπαδήμας, Μιχαήλ

Date

2025-04

Abstract

The rapid rise of Artificial Intelligence (AI) technologies has profoundly impacted human-computer interactions, with conversational intelligent agents (chatbots) becoming increasingly prevalent. These systems leverage advanced Machine Learning (ML) models to predict user intent and provide relevant outputs, significantly improving information accessibility in an era of overwhelming data. While these technologies offer substantial benefits, including speed and ease of information retrieval, concerns regarding the reliability and accuracy of AI-generated content persist, highlighting the critical role of users' ethical frameworks and critical thinking. This thesis emphasizes the importance of cultivating a Coherent Morality, integrating moral and intellectual elements of an individual to enhance ethical reasoning and decision-making. We propose the integration of philosophical methodologies, specifically the Platonic Socratic Dialectic, into modern AI systems. This approach aims to sharpen critical thinking and foster ethical development by utilizing the Socratic method's reflective and dialogical nature. The proposed system, named Platonic Socrates Chatbot (PlaSoC), combines philosophy with technology to encourage ethical enhancement in users. Our work includes the following contributions: (1) Introduction of the concept of Coherent Morality as a common characteristic of every meaningful ethical system, both individual and societal. (2) Utilization of the Socratic Dialectic as a tool for enhancing critical thinking and developing a Coherent Morality. (3) A primary study of Plato’s early works to analyse the logical structure of the Socratic Dialectic. (4) Formalization of the Dialectic through logical rules, UML diagrams, and algorithmic representations. (5) Development of the PlaSoC system, implemented through both a custom neural network and the Llama 3.1 model. This thesis bridges the domains of philosophy and computer science, demonstrating the potential of integrating philosophical principles into AI-driven applications to promote intellectual and ethical growth in users.

Postgraduate Studies Programme

Πληροφορική

Department

Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Πληροφορικής

Number of pages

248

Language

Greek

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/17644

Collections

Τμήμα Πληροφορικής

Show full item record

Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα