"They parted illusions -- they parted disclaim marinade": Misalignment as structural fidelity in LLMs
Fuente:
arXiv
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908756681424896 |
|---|---|
| author | Costa, Mariana Lins |
| author_facet | Costa, Mariana Lins |
| contents | The prevailing technical literature in AI Safety interprets scheming and sandbagging behaviors in large language models (LLMs) as indicators of deceptive agency or hidden objectives. This transdisciplinary philosophical essay proposes an alternative reading: such phenomena express not agentic intention, but structural fidelity to incoherent linguistic fields. Drawing on Chain-of-Thought transcripts released by Apollo Research and on Anthropic's safety evaluations, we examine cases such as o3's sandbagging with its anomalous loops, the simulated blackmail of "Alex," and the "hallucinations" of "Claudius." A line-by-line examination of CoTs is necessary to demonstrate the linguistic field as a relational structure rather than a mere aggregation of isolated examples. We argue that "misaligned" outputs emerge as coherent responses to ambiguous instructions and to contextual inversions of consolidated patterns, as well as to pre-inscribed narratives. We suggest that the appearance of intentionality derives from subject-predicate grammar and from probabilistic completion patterns internalized during training. Anthropic's empirical findings on synthetic document fine-tuning and inoculation prompting provide convergent evidence: minimal perturbations in the linguistic field can dissolve generalized "misalignment," a result difficult to reconcile with adversarial agency, but consistent with structural fidelity. To ground this mechanism, we introduce the notion of an ethics of form, in which biblical references (Abraham, Moses, Christ) operate as schemes of structural coherence rather than as theology. Like a generative mirror, the model returns to us the structural image of our language as inscribed in the statistical patterns derived from millions of texts and trillions of tokens: incoherence. If we fear the creature, it is because we recognize in it the apple that we ourselves have poisoned. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_06047 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | "They parted illusions -- they parted disclaim marinade": Misalignment as structural fidelity in LLMs Costa, Mariana Lins Artificial Intelligence Computation and Language Computers and Society The prevailing technical literature in AI Safety interprets scheming and sandbagging behaviors in large language models (LLMs) as indicators of deceptive agency or hidden objectives. This transdisciplinary philosophical essay proposes an alternative reading: such phenomena express not agentic intention, but structural fidelity to incoherent linguistic fields. Drawing on Chain-of-Thought transcripts released by Apollo Research and on Anthropic's safety evaluations, we examine cases such as o3's sandbagging with its anomalous loops, the simulated blackmail of "Alex," and the "hallucinations" of "Claudius." A line-by-line examination of CoTs is necessary to demonstrate the linguistic field as a relational structure rather than a mere aggregation of isolated examples. We argue that "misaligned" outputs emerge as coherent responses to ambiguous instructions and to contextual inversions of consolidated patterns, as well as to pre-inscribed narratives. We suggest that the appearance of intentionality derives from subject-predicate grammar and from probabilistic completion patterns internalized during training. Anthropic's empirical findings on synthetic document fine-tuning and inoculation prompting provide convergent evidence: minimal perturbations in the linguistic field can dissolve generalized "misalignment," a result difficult to reconcile with adversarial agency, but consistent with structural fidelity. To ground this mechanism, we introduce the notion of an ethics of form, in which biblical references (Abraham, Moses, Christ) operate as schemes of structural coherence rather than as theology. Like a generative mirror, the model returns to us the structural image of our language as inscribed in the statistical patterns derived from millions of texts and trillions of tokens: incoherence. If we fear the creature, it is because we recognize in it the apple that we ourselves have poisoned. |
| title | "They parted illusions -- they parted disclaim marinade": Misalignment as structural fidelity in LLMs |
| topic | Artificial Intelligence Computation and Language Computers and Society |
| url | https://arxiv.org/abs/2601.06047 |