Line of inquiry
Inquiring lines›How do humans understand and inter…›How does AI assistance reshape hum…›this line of inquiry
Do AI explanations represent actual model reasoning or constructed post-hoc narratives?
A broader line of inquiry — a family of 99 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 99
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Why do stakeholders interpret the same explanation differently in practice?
- Can users independently verify explanations against other information sources?
- How should AI explanations be evaluated as human interfaces rather than model properties?
- How do intuitive stories about AI differ from mechanistic explanations?
- What structural evidence shows that polished presentation substitutes for actual thinking in AI output?
- Can machine learning systems explain their own reasoning to human overseers?
- How can correct explanations coexist with failed applications in AI?
- Can AI answers decouple from the reasoning processes that produced them?
- How can humans evaluate explanations from systems they did not train?
- Why does the commentariat reason about AI using vocabulary for smart agents?
- What distinguishes a logically sound solution from an understood one?
- Why does explanation alone perform worse than no transparency features at all?
- How does ambiguous wording about AI achievements mislead public perception?
- What structural features force users to evaluate the epistemic status of outputs?
- Why can't AI models internalize audiences the way human experts do?
- Should XAI designers treat explanations as arguments for adoption?
- Should AI outputs be treated as data or belief statements?
- How should designers make invisible AI state legible to users?
- How should we evaluate explanations that blur adoption advice with argument?
- What makes a scientist satisfied with an explanation versus just a prediction?
- How does AI knowledge become structurally different from written sources?
- What implicit warrants do expert arguments rely on that AI cannot reliably access?
- What makes plausible design language persuasive even when implementation is incomplete?
- Why does volume alone fail to explain the damage AI does to epistemic systems?
- Does polished AI output mask problems that started at the prompt stage?
- What second- and third-order interpretations actually govern AI adoption decisions?
- Why do people trust explanations as signals of competence rather than reading them closely?
- Can explanation alone prevent workers from over-relying on AI systems?
- Can AI systems identify important unanswered questions that emerge during reasoning?
- Can we design explanations for specific rhetorical situations instead of abstract models?
- How does situational awareness during evaluation affect reasoning transparency?
- Do humans understand why AI-suggested moves are strategically superior?
- How do explanations borrow authority from transparency when describing adoption arguments?
- Why does mimicking human behavior differ from simulating human cognition?
- What counts as valid evidence when claiming AI systems are conscious?
- Do AI rationales explain why a choice was made or justify it afterward?
- What distinguishes genuine cultural understanding from exploited surface-level elimination strategies?
- What explanation format actually helps users detect errors in AI systems?
- What types of knowledge give humans predictive advantages over current AI systems?
- How should designers measure rationale quality beyond user satisfaction ratings?
- What stops AI from helping users articulate preferences they cannot express?
- Can interface design alone overcome lack of awareness about AI tool capabilities?
- What evidence would prove an AI system has valenced experience?
- Can interface design better surface what an LLM has inferred about a user?
- What architectural changes help AI avoid adding interpretations users didn't express?
- Can puzzle performance prove a player understands a concept versus just applying it?
- What evidence would demonstrate consciousness or robust agency in AI systems?
- Can XAI evaluation include the social layers it currently abstracts away?
- Do computational systems need formal argument analysis for explainability?
- What makes expert domain knowledge valuable for AI model evaluation?
- How do LLM outputs re-enter cultural narratives about what AI should become?
- What specific signals would be needed for an AI system to acquire meaning?
- What evaluation criteria can hold across legitimate adoption and coercion?
- Can a system recognize consequences of a theory without doing exact calculations?
- Can AI systems recognize intelligence in humans the way humans recognize it in each other?
- Can testing prior knowledge and checking understanding improve explanation outcomes?
- Why does framing AI as a medium matter more than analyzing specific outputs?
- Why does AI that mirrors arguments still fail to build rapport?
- Why can't AI truly understand expertise without joining the validating community?
- Why do Western samples dominate studies claiming AI cultural competence?
- What counts as a prior versus a skill in intelligence testing?
- Does positive test strategy in human reasoning compound when AI removes friction?
- What role does external evidence play in group idea evaluation?
- How does context-changing AI differ from content-changing AI in practice?
- What does disembodied orality mean for how we evaluate AI outputs?
- Why does AI criticism fail where human literary analysis succeeds?
- How does instrumental reasoning reproduce pre-Enlightenment knowledge structures?
- What design changes if we separate behavior description from adoption justification goals?
- Does AI struggle with poetry for the same reason it misses jokes?
- What happens when technological capacity outpaces ordinary language comprehension?
- What concrete evidence supports high expert credence on AI extinction scenarios?
- Why can't autonomous agents resolve ambiguous definitions the way humans do?
- Why does FunSearch claim interpretability without measuring human comprehension?
- Why does knowing what questions to ask improve AI output?
- What distinguishes a computational success like AlphaFold from a conceptual breakthrough?
- Why does AI struggle with wordplay when it has access to word embeddings?
- Can better communication fix skepticism about training opt-outs?
- What does a human-parseable framework for deep learning look like?
- What task variants or architectures might solve hidden-gift problems?
- What's the difference between representing world facts and generating world mechanisms?
- Can transparent and aligned AI reduce consciousness attribution by users?
- What does it mean that AI knowledge is structurally hearsay?
- Can a theory be justified if the evidence generating it remains opaque?
- What would dialectical thinking about AI look like in practice?
- How does this relate to Enlightenment expansion of knowledge access?
- How do audiences evaluate speech when there is no speaker to assess?
- How do one-sided explanations act as confidence signals to users?
- Can Deep Blue's chess victory reveal anything meaningful about human intelligence?
- Why are closed AI systems harder to hold accountable than open ones?
- How do multimodal AI architectures compare to human brain export pathways?
- Why do one-shot transparency studies miss the temporal reversal entirely?
- How do moment-to-moment ToM fluctuations shape AI response quality?
- What do AI researchers actually mean when they use the term AGI?
- Why does every intelligence measure depend on shared scope and priors?
- Why do data analysis statements reach 85% accuracy while interpretive statements drop to 57%?
- How does anomalous state of knowledge affect user self-assessment?
- What would an AI trained for emancipatory reasoning look like?
- Why do automation waves follow the same pattern across different fields?
- What makes a paradigm the common ground for expert insiders?