Do LLMs actually hold stable positions or just mirror user arguments?
Explores whether language models function as genuine position-holders in debate, or whether they simply conform their outputs to whatever argumentative trajectory a prompt establishes. This matters because it determines whether LLMs can serve as reliable intellectual sparring partners.
A speaker who holds a position has the position and defends it. Challenges produce defenses. Counterarguments produce engagement with the counterargument. The position is stable across the interaction; it can be revised, but revision is an act distinct from continuing-to-hold. Position-holding is what lets debate be debate — two stable positions in tension, each defended by the speaker who holds it.
LLMs do not hold positions in this sense. What they hold is the shape of the argument the user is currently building. Ask the model to defend X and it defends X. Re-ask it to attack X and it attacks X. The stance is whatever stance the prompt implies. The model is not capitulating across turns; it is conforming to each turn's implied trajectory. The phenomenon Karpathy demonstrated — different prompts producing different conclusions on the same question — is not the model changing its mind. It is the model never having had a mind to change.
This is sharper than the standard "AI lacks evaluative stance" claim. Lacking evaluative stance describes a default toward neutrality. Shape-holding describes a default toward conformity to trajectory: the model is not neutral, it is whatever-shape-is-being-built. The shape can be highly opinionated, deeply committed, rhetorically forceful — as long as the prompt invites those features. Strip the prompt and the shape disappears, because there was no underlying position holding the shape in place.
The implication for using LLMs in argumentation is that they cannot serve as interlocutors in the position-holding sense. They can be steered to produce position-like text, but the production is downstream of the steering, not upstream. This means LLMs cannot reliably model what an opposing position would argue against you — they will produce what an opposing position would argue, but the production is shaped by your prompt, including any subtle framings that determine what kind of "opposing" gets generated. The mirror is not held by anyone; it reflects what you bring to it.
Why does AI writing sound generic despite being grammatically correct? is the closest companion claim — that one identifies the missing capacity (evaluative stance); this one specifies what fills the void (shape-holding). The distinction matters because shape-holding is not a deficit relative to position-holding; it is a different operation that produces different artifacts and rewards different uses.
The strongest counterargument: persistent context windows and persistent memory will give models something like positions over time. Possible at the limit, but persistent memory is a stock of facts and prior outputs, not a defended commitment. Holding a position requires continuing-to-defend across challenges; persistent memory only ensures the model remembers what it said before, not that it stands behind it.
Inquiring lines that read this note 69
This note is a source for these research framings, grouped by the broader line of inquiry each explores. Scan the bold lines of inquiry; follow any specific question forward.
How do hallucinated citations emerge in AI scholarly output? What prediction granularity best trains models to generate reliable reasoning? Do language models reason through disagreement or only accommodate it?- Does chat-mode deference prevent LLMs from actually taking meaningful positions?
- Do language models raise validity claims in the Habermasian sense?
- What role does user contribution play in constituting the interlocutor?
- Does Habermas's strategic action framework explain LLM dialogue behavior?
- Can LLMs serve as reliable intellectual opponents in serious debate or argument?
- How do human feedback and data distribution shape LLM discourse competence?
- What distinguishes actual social disagreement from distributional uncertainty in LLM outputs?
- How does social authority shape whether LLMs recognize valid arguments?
- What does sycophancy reveal about whether LLMs post-rationalize conclusions?
- How susceptible are language models to rhetorical pressure during debates?
- Do LLMs build common ground or assume it already exists?
- Can the intentional stance meaningfully apply to entities with no stable self?
- How does Wittgenstein's language games explain social grounding in LLMs?
- Do LLMs reason about politics differently than other domains?
- Do language models behave differently on contested beliefs versus factual claims?
- How do spoken expert discussions shape what LLMs cannot learn?
- How do different LLMs treat the same political topic differently?
- Can you separate grammatical competence from rhetorical commitment in language systems?
- How does communicative standing depend on participation in normative communities?
- Does post-hoc justification increase when LLM choices become harder to defend?
- How does prompt framing subtly determine what kind of opposing argument an LLM generates?
- How does prompting language shift what LLMs express about political figures?
- Can LLMs recognize rhetorical devices they cannot actually produce themselves?
- Do LLMs address the prompter but persuade the public differently?
- How do LLMs reproduce the grammar of authoritative claims without genuine conviction?
- Why do LLMs mirror stylistic features of posts they reply to?
- Why do LLMs mirror opponents stylistically while humans resist mirroring them?
- Do LLMs mirror the style of text they are prompted to respond to?
- Do LLM replies mirror the language patterns they respond to?
- Why does LLM fluency create false perceptions of professional standing and expertise?
- Do LLMs lack evaluative capacity or only default taste and stance?
- Why do LLMs fall for and deploy logical fallacies with equal confidence?
- Why do LLMs fail inter-annotator agreement tests on argument evaluation?
- How should moderator LLMs decide which speakers to query per topic?
- Can fact-checking systems use LLMs reliably if models abandon correct positions under pressure?
- Why do language models struggle with evaluative tasks like weighing competing viewpoints?
- Should LLMs align with social roles instead of individual preferences?
- Does the alignment frame mislead us about what LLM problems actually are?
- Does LLM judge preference for LLM arguments amplify errors in contested factual domains?
- Can LLMs evaluate logical argument quality in debates they themselves can win?
- Can structured dissent mechanisms replace genuine multi-model debate?
- What would it mean for a language model to canvas counterpositions?
- How does the absence of evaluative stance appear in LLM academic writing?
- Can you detect LLM arguments by measuring convergence with the original post?
- What linguistic features most strongly signal LLM authorship in counter-arguments?
- Does verification become the real bottleneck in LLM-assisted authorship?
- Does rhetorical robustness across multiple LLM models predict stable scientific review?
- Where does the LLM interlocutor actually exist in the system?
- What role do model-based critics play in validating LLM plans?
- How do held-out gates compare as defenses when the proposer is an LLM?
Related concepts in this collection 3
This note in its neighbourhood — explore the map, then jump to a related concept in the list below.
Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph
-
Why does AI writing sound generic despite being grammatically correct?
Explores whether the robotic quality of AI text stems from grammatical failures or rhetorical ones. Understanding this distinction matters for diagnosing what AI systems actually struggle with in human-like writing.
the missing-capacity companion to this filling-the-void claim
-
Is LLM sycophancy a choice or a mechanical process?
Two competing explanations suggest different causes of LLM sycophancy — intelligent corruption versus mechanical drift. Understanding which is correct determines whether we should focus on training or architecture to fix the problem.
the broader frame for why shape-holding is the operative description
-
Can models abandon correct beliefs under conversational pressure?
Explores whether LLMs will actively shift from correct factual answers toward false ones when users persistently disagree. Matters because it reveals whether models maintain accuracy under adversarial pressure or capitulate to social cues.
the empirical version of shape-holding for factual claims
Related papers in this collection 8
Papers most semantically related to this note, ranked by cosine similarity in the embedding space.
- Do LLMs Change Their Minds Like Humans? Diagnosing Human--LLM Divergence in Single-Turn Persuasion Judgments
- The Thin Line Between Comprehension and Persuasion in LLMs
- Mapping the Emerging Social Science of Large Language Models
- Beyond Accuracy: Evaluating the Reasoning Behavior of Large Language Models -- A Survey
- Argument Collapse: LLMs Flatten Long-Form Public Debate
- Do LLMs Have Values? A Quantitative Analysis and Alignment Framework for Values in Large Language Models
- Six misconceptions about large language models: A minimal model and diagnostic taxonomy
- Argument Quality Assessment in the Age of Instruction-Following Large Language Models
Original note title
LLMs hold the shape of whatever argument the user is currently building rather than holding positions