Here's Charlie! Realising the Semantic Web vision of Agents in the age of LLMs

TL;DR

This paper explores developing trustworthy semi-autonomous AI agents for online interactions, enabling user control over data and decisions while leveraging LLMs for natural language communication and safety guarantees.

Contribution

It introduces a framework combining rules and LLMs for trustworthy, user-controlled semi-autonomous agents, demonstrated through a personal assistant use case.

Findings

Demonstrated a prototype personal assistant with safety guarantees.

Showed effective user-agent dialogue for trust and control.

Integrated rules and LLMs for reliable agent behavior.

Abstract

This paper presents our research towards a near-term future in which legal entities, such as individuals and organisations can entrust semi-autonomous AI-driven agents to carry out online interactions on their behalf. The author's research concerns the development of semi-autonomous Web agents, which consult users if and only if the system does not have sufficient context or confidence to proceed working autonomously. This creates a user-agent dialogue that allows the user to teach the agent about the information sources they trust, their data-sharing preferences, and their decision-making preferences. Ultimately, this enables the user to maximise control over their data and decisions while retaining the convenience of using agents, including those driven by LLMs. In view of developing near-term solutions, the research seeks to answer the question: "How do we build a trustworthy and reliable network of semi-autonomous agents which represent individuals and organisations on the Web?". After identifying key requirements, the paper presents a demo for a sample use case of a generic personal assistant. This is implemented using (Notation3) rules to enforce safety guarantees around belief, data sharing and data usage and LLMs to allow natural language interaction with users and serendipitous dialogues between software agents.