How does AI-assisted work reshape how people see their own abilities?
When users delegate tasks to AI, do they unknowingly integrate the system's outputs into their sense of personal competence? This explores whether AI interaction produces a specific form of self-perception distortion distinct from trust or effort issues.
The literature on AI interaction risks has three well-established constructs that the LLM Fallacy must be distinguished from, because conflating them produces wrong interventions.
Hallucination is a system-level failure: the model produces incorrect or fabricated information. The LLM Fallacy is independent of output correctness — it persists regardless of whether generated content is accurate or erroneous, because it operates at the level of attribution rather than epistemic validity. A user can experience the LLM Fallacy even when every AI output they receive is perfectly correct.
Automation bias involves over-reliance on system outputs in decision-making. The focus is on task execution: users follow system recommendations without sufficient scrutiny. The LLM Fallacy extends beyond reliance into capability attribution — it is not about trusting the system too much but about believing you could produce the output yourself.
Cognitive offloading involves delegating mental effort to external systems. The focus is on effort management: users outsource cognitive work to reduce load. The LLM Fallacy concerns how the outsourced outputs are integrated into self-perception — not the delegation itself but the failure to update one's self-model to account for the delegation.
The practical consequence of the distinction: interventions for hallucination (better retrieval, factual grounding) do not address the LLM Fallacy. Interventions for automation bias (forcing manual verification) partially address it but miss the self-perception layer. Interventions for cognitive offloading (forcing engagement) help but are framed as effort problems rather than identity problems. The LLM Fallacy requires interventions that make the human-machine contribution boundary salient — not just accurate outputs or forced engagement but structural transparency about who did what.
Inquiring lines that read this note 42
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.
Does self-reflection enable models to reliably correct their errors? How can humans calibrate appropriate trust in AI systems?- How does outcome feedback change beliefs about AI versus human partner reliability?
- What distinguishes misattributed social role from misattributed competence in AI trust failures?
- Which workplace tasks see productivity gains when AI and users align?
- What individual differences predict who benefits from AI partnership?
- How does rising AI capability change what users expect from their tools?
- Why does AI-improved task performance fail to transfer to independent work?
- Does outsourcing tasks to AI reduce opportunities for skill development?
- Does democratizing AI access actually improve or impair human skill development?
- Does broader AI access empower people or gradually disempower human agency?
- Why do interventions for hallucination or automation bias fail to address capability misattribution?
- How does capability differ from what workers actually want from AI?
- How do interpersonal skills reshape task importance as automation increases?
- Do workers become dependent on AI when they stop using it for the same task?
- How does AI reliance change professional judgment and autonomy?
- Is the shift toward interpersonal skills a permanent role or a temporary phase before full automation?
- Why do expert roles shift when AI generates rather than humans?
- Why do users feel more competent when their actual capability is declining?
- What mechanisms make users misattribute AI outputs as their own competence?
- Why do users believe they produced independent competence when they actually used AI assistance?
- Why do people misattribute AI outputs as evidence of their own skill?
- How does opaque AI processing distort users' perception of their contribution?
- Why does polished AI output feel like evidence of user skill?
- What happens when users mistake AI assistance for their own competence?
- How much does autonomous action without prompting affect user perception?
- What distinguishes perception contribution from decision authority in collaboration?
- Does AI assistance actually reduce neural processing and brain connectivity over time?
- How does anomalous state of knowledge affect user self-assessment?
- What happens to the brain when people rely on AI assistance repeatedly?
- Can users tell the difference between their own thinking and AI contribution?
- How does AI assistance affect human cognitive development over time?
- How does AI reliance connect to the gap between perceived and actual competence?
- How do evaluation systems shift power between humans and AI outputs?
- Can users adapt their competencies to match how AI actually operates?
Related concepts in this collection 2
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
-
Do AI-assisted outputs fool users about their own skills?
When people use AI tools to produce high-quality work, do they mistakenly believe they personally possess the skills that generated it? This matters because such misattribution could mask genuine skill loss and prevent corrective action.
the parent concept
-
Why do people trust AI outputs they shouldn't?
When do human cognitive shortcuts fail in AI interaction? Three compounding traps—treating statistical patterns as facts, mistaking fluency for understanding, and avoiding disagreement—may explain systematic overreliance across languages and contexts.
Rose-Frame's Trap 2 (mistaking fluency for understanding) is a component of the LLM Fallacy, not the whole phenomenon
Related papers in this collection 8
Papers most semantically related to this note, ranked by cosine similarity in the embedding space.
- The LLM Fallacy: Misattribution in AI-Assisted Cognitive Workflows
- Machine Bullshit: Characterizing the Emergent Disregard for Truth in Large Language Models
- Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
- Language Models Learn to Mislead Humans via RLHF
- Beyond Hallucinations: The Illusion of Understanding in Large Language Models
- The Moral Turing Test: Evaluating Human-LLM Alignment in Moral Decision-Making
- Beyond Accuracy: The Role of Calibration in Self-Improving Large Language Models
- Evaluating Large Language Models in Theory of Mind Tasks
Original note title
the LLM Fallacy is distinct from hallucination automation bias and cognitive offloading — it operates at the level of self-perception not task execution or system reliability