# kaal:claim:6421319-023

**Claim.** Probabilistic language models hallucinate because they are trained to predict statistically likely token sequences rather than to verify propositional truth, so the error is intrinsic to the substrate: plausibility and truth are orthogonal properties in high dimensional token space.

**Type.** mechanism  **Support.** argued

**Holds when.**

- applies to purely probabilistic architectures

**Source quote.**

> Probabilistic large language models hallucinate because they are trained to predict statistically likely token sequences, not to verify propositional truth.

**From.** Wulf A. Kaal, *The Collapse of Scarcity Economics* (2026), V.D. The Hallucination Objection: Residual Imperfection and the Persistence of Signaling, page 21

**Cite as.** Wulf A. Kaal, The Collapse of Scarcity Economics (2026). SSRN: https://ssrn.com/abstract=6421319

**Verify.** sha256 of source PDF `0f4f68e56ee48bb8b2d73d3ab7fd96fa572e76c195694ffa740360704c1b172d` at https://raw.githubusercontent.com/wulfkaal/Academic-Papers/main/papers/pdf/Kaal%20-%202026%20-%20The%20Collapse%20of%20Scarcity%20Economics.pdf

**Failure mode.** Plausibility Truth Orthogonality  (family: ai-model-and-training-failure)

**Topics.** institutional-design

**Keywords.** hallucination, language-models, architecture, truth-verification

**Canonical form.** This markdown file is the canonical hashed representation of the claim. Its sha256 is the content hash used for attestation.