Beyond Chat: a Framework for LLMs as Human-Centered Support Systems

TL;DR

This paper introduces a role-based framework for human-centered large language model support systems, emphasizing design principles, evaluation metrics, and addressing risks to promote responsible deployment in sensitive, guidance-oriented contexts.

Contribution

It proposes a novel role-based framework for LLMs as human support systems, compares real-world deployments, and outlines comprehensive design principles and evaluation metrics.

Findings

Identifies key design principles: transparency, personalization, guardrails, memory with privacy, empathy, and reliability.

Extends evaluation metrics beyond accuracy to include trust, engagement, and long-term outcomes.

Analyzes risks like over-reliance, hallucination, bias, privacy issues, and access inequality.

Abstract

Large language models are moving beyond transactional question answering to act as companions, coaches, mediators, and curators that scaffold human growth, decision-making, and well-being. This paper proposes a role-based framework for human-centered LLM support systems, compares real deployments across domains, and identifies cross-cutting design principles: transparency, personalization, guardrails, memory with privacy, and a balance of empathy and reliability. It outlines evaluation metrics that extend beyond accuracy to trust, engagement, and longitudinal outcomes. It also analyzes risks including over-reliance, hallucination, bias, privacy exposure, and unequal access, and proposes future directions spanning unified evaluation, hybrid human-AI models, memory architectures, cross-domain benchmarking, and governance. The goal is to support responsible integration of LLMs in sensitive settings where people need accompaniment and guidance, not only answers.