The Start Button Problem: a basis for human responsibility in artificial intelligence computation

TL;DR

This paper discusses the boundaries of AI autonomy, emphasizing human responsibility by analyzing the Start Button Problem and its implications for AI control and dependency.

Contribution

It introduces the Start Button Problem as a new thought experiment to examine AI autonomy and human responsibility in AI systems.

Findings

AI's need for human activation underscores its dependency.

The Start Button Problem challenges assumptions of AI as independent agents.

Human responsibility extends to AI activation and purpose setting.

Abstract

Recent advancements in artificial intelligence have reopened the question about the boundaries of AI autonomy, particularly in discussions around artificial general intelligence and its potential to act independently across varied purposes. This paper explores these boundaries through the analysis of the Alignment Research Center experiment on GPT-4 and introduces the Start Button Problem, a thought experiment that examines the origins and limits of AI autonomy. By examining the thought experiment and its counterarguments, it becomes clear that in its need for human activation and purpose definition lies the AI's inherent dependency on human-initiated actions, challenging the assumption of AI as an intelligent agent. Finally, the paper addresses the implications of this dependency on human responsibility, questioning the measure of the extension of human responsibility when using AI systems.