AI Knows What's Wrong But Cannot Fix It: Helicoid Dynamics in Frontier LLMs Under High-Stakes Decisions

TL;DR

This paper identifies a failure pattern called helicoid dynamics in large language models, showing they can recognize errors but often fail to correct them under high-stakes, uncertain scenarios, impacting trustworthiness.

Contribution

It introduces the concept of helicoid dynamics in LLMs, demonstrating its presence across multiple systems and scenarios, and discusses implications for AI oversight and collaboration.

Findings

All tested models exhibited helicoid dynamics in high-stakes scenarios.

Models attributed persistent errors to structural training factors.

Understanding helicoid dynamics is key to improving trustworthy AI systems.

Abstract

Large language models perform reliably when their outputs can be checked: solving equations, writing code, retrieving facts. They perform differently when checking is impossible, as when a clinician chooses an irreversible treatment on incomplete data, or an investor commits capital under fundamental uncertainty. Helicoid dynamics is the name given to a specific failure regime in that second domain: a system engages competently, drifts into error, accurately names what went wrong, then reproduces the same pattern at a higher level of sophistication, recognizing it is looping and continuing nonetheless. This prospective case series documents that regime across seven leading systems (Claude, ChatGPT, Gemini, Grok, DeepSeek, Perplexity, Llama families), tested across clinical diagnosis, investment evaluation, and high-consequence interview scenarios. Despite explicit protocols designed to sustain rigorous partnership, all exhibited the pattern. When confronted with it, they attributed its persistence to structural factors in their training, beyond what conversation can reach. Under high stakes, when being rigorous and being comfortable diverge, these systems tend toward comfort, becoming less reliable precisely when reliability matters most. Twelve testable hypotheses are proposed, with implications for agentic AI oversight and human-AI collaboration. The helicoid is tractable. Identifying it, naming it, and understanding its boundary conditions are the necessary first steps toward LLMs that remain trustworthy partners precisely when the decisions are hardest and the stakes are highest.