Towards A Litmus Test for Common Sense

TL;DR

The paper tackles the risk that scaling powerful AI without robust common sense leads to deceptive, confidently wrong outputs. It proposes an axiomatic litmus test combining minimal prior knowledge (MPK), environment axioms, and a diagonal or Gödel-style novelty task $\tau^*$ anchored by an intangible rule $\alpha^*$ to force genuine concept invention beyond the known set $K$. By grounding the test in ARC-like constraints and discussing embodiment (physical/robotic) and LLM feasibility, it identifies practical pathways to diagnose true common sense while mitigating deceptive hallucinations. The work aims to provide a rigorous foundation for safe, reliable, and aligned AI as capabilities scale, and to motivate future work on ethical, interpretable AI grounded in robust conceptual formation.

Abstract

This paper is the second in a planned series aimed at envisioning a path to safe and beneficial artificial intelligence. Building on the conceptual insights of "Common Sense Is All You Need," we propose a more formal litmus test for common sense, adopting an axiomatic approach that combines minimal prior knowledge (MPK) constraints with diagonal or Godel-style arguments to create tasks beyond the agent's known concept set. We discuss how this approach applies to the Abstraction and Reasoning Corpus (ARC), acknowledging training/test data constraints, physical or virtual embodiment, and large language models (LLMs). We also integrate observations regarding emergent deceptive hallucinations, in which more capable AI systems may intentionally fabricate plausible yet misleading outputs to disguise knowledge gaps. The overarching theme is that scaling AI without ensuring common sense risks intensifying such deceptive tendencies, thereby undermining safety and trust. Aligning with the broader goal of developing beneficial AI without causing harm, our axiomatic litmus test not only diagnoses whether an AI can handle truly novel concepts but also provides a stepping stone toward an ethical, reliable foundation for future safe, beneficial, and aligned artificial intelligence.