mlsanityData to 14 Oct 2026

Many-body Tipping Dynamics of ChatGPT-like AIs

arXiv 2607.25279 ↗First posted 28 Jul 2026cs.AI cond-mat.dis-nn math-phcited by 0

Not on the map. Why?

Abstract (arXiv)

Why do ChatGPT-like AIs, despite major architectural and training differences, unexpectedly tip to undesirable content (e.g. harmful, misleading, repetitive) even under deterministic greedy decoding? We show that a broad class of such tippings is caused by the many-body interactions between tokens (spins) as they cross the finite-layer system. Tipping emerges as a dynamical first passage process between competing output basins. Attention disorder controls the transport toward, away from, or along the basins' boundary. A few-basin reduction yields a closed finite-layer threshold, whose coarse-grained predictions show good agreement across ChatGPT-like families. These results suggest that a broad class of AI failures represents 'foreseeable engineering risk' rather than inherently unpredictable behavior, with important implications for legal and societal assessments of AI harm.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 22 of its 42 references that are in the map. First posted on arXiv 28 Jul 2026.

Builds on

22 of 42 references.

References · 22

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  3. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  4. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  5. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,420
  6. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,336
  7. 2024Refusal in Language Models Is Mediated by a Single Directioncited by 1,032
  8. 2023Progress measures for grokking via mechanistic interpretabilitycited by 839
  9. 2021Attention is Not All You Need: Pure Attention Loses Rank Doubly Exponentially with Depthcited by 419
  10. 2023The emergence of clusters in self-attention dynamicscited by 140
  11. 2020A Theoretical Analysis of the Repetition Problem in Text Generationcited by 83
  12. 2026Verbalizable Representations Form a Global Workspace in Language Modelscited by 75
10 more
  1. 2025The Curse of Depth in Large Language Modelscited by 62
  2. 2025The Mean-Field Dynamics of Transformerscited by 30
  3. 2026Characterizing Delusional Spirals through Human-LLM Chat Logscited by 23
  4. 2024Phase transition in large language models and the criticality of natural languagescited by 13
  5. 2023Where's the Liability in Harmful AI Speech?cited by 12
  6. 2025Phase Transitions in Large Language Models and the O(N) Modelcited by 12
  7. 2025From Condensation to Rank Collapse: A Two-Stage Analysis of Transformer Training Dynamicscited by 10
  8. 2025Shadows in the Attention: Contextual Perturbation and Representation Drift in the Dynamics of Hallucination in LLMscited by 9
  9. 2026Do Thinking Tokens Help with Safety?cited by 3
  10. 2025Testing the spin-bath view of self-attention: A Hamiltonian analysis of GPT-2 Transformercited by 2

Built on by

Not cited yet.