Hermeneutic Sovereignty under Generative AI: A Humanities Framework for Protecting Human Meaning-Making from Interpretive Foreclosure

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The NIST AI Risk Management Framework provides a practical structure for identifying and managing risks across the AI lifecycle, and its generative AI profile extends that orientation to risks associated with generative systems.

“The NIST AI Risk Management Framework, for example, provides a practical structure for identifying and managing risks across the AI lifecycle, while its generative AI profile extends that orientation to risks associated with generative systems (Autio et al., 2024;Tabassi, 2023).”

• cites Autio et al., 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Tabassi, 2023 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Ethical frameworks such as AI4People identify autonomy, justice, explicability and other principles that should inform responsible AI.

“Ethical frameworks such as AI4People identify autonomy, justice, explicability and other principles that should inform responsible AI (Floridi et al., 2018).”

• cites Floridi et al., 2018 sourceSupported

Classified as probably not a substantial claim.

Milano and Prunkl (2025) argue that algorithmic profiling can generate hermeneutical injustice through epistemic fragmentation, making it harder for individuals to identify and conceptualise shared harms.

“Milano and Prunkl (2025) argue that algorithmic profiling can generate hermeneutical injustice through epistemic fragmentation, making it harder for individuals to identify and conceptualise shared harms.”

• cites Milano and Prunkl (2025) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Recent AI scholarship extends epistemic injustice concerns to algorithmic profiling and generative systems.

“Recent AI scholarship extends these concerns to algorithmic profiling and generative systems (Kay et al., 2024;Milano & Prunkl, 2025).”

• cites Kay et al., 2024 sourceSupported
• cites Milano & Prunkl, 2025 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Kay et al. (2024) develop the idea of generative algorithmic epistemic injustice, including hermeneutical ignorance and access injustice within AI-mediated knowledge ecosystems.

“Kay et al. (2024) develop the idea of generative algorithmic epistemic injustice, including hermeneutical ignorance and access injustice within AI-mediated knowledge ecosystems.”

• cites Kay et al. (2024) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The NIST risk management approach is lifecycle-oriented and deliberately broad.

“The NIST riskmanagement approach is lifecycle-oriented and deliberately broad (Autio et al., 2024;Tabassi, 2023).”

• cites Autio et al., 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Tabassi, 2023 sourceSupported

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Risk-management frameworks provide valuable organisational scaffolding.

“risk-management frameworks provide valuable organisational scaffolding (Autio et al., 2024;Tabassi, 2023)”

• cites Autio et al., 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Tabassi, 2023 sourceSupported

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Autonomy, justice and explicability remain essential.

“Autonomy, justice and explicability remain essential (Floridi et al., 2018)”

• cites Floridi et al., 2018 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Erickson and Gregory (2025) make a related jurisprudential argument that algorithmic demands for specification can undermine the productive vagueness of legal concepts and transfer interpretive authority to extra-legal processes.

“Erickson and Gregory (2025) make a related jurisprudential argument that algorithmic demands for specification can undermine the productive vagueness of legal concepts and transfer interpretive authority to extra-legal processes.”

• cites Erickson and Gregory (2025) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Sakabe (2026) uses interpretive sovereignty to describe human creative subjectivity in an analysis of artistic appropriation in the era of generative AI.

“Sakabe (2026) uses interpretive sovereignty to describe human creative subjectivity in an analysis of artistic appropriation in the era of generative AI.”

• cites Sakabe (2026) sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

The humanities problem of generative AI begins where assistance becomes interpretation.

“The humanities problem begins where assistance becomes interpretation.”

Probably not a substantial claim: may not pass S4 (based).

Generative AI increasingly mediates not only what institutions decide, but how institutions interpret people.

“Generative artificial intelligence increasingly mediates not only what institutions decide, but how institutions interpret people.”

Probably not a substantial claim: may not pass S2 (checkable).

The risk of AI-mediated interpretation is interpretive foreclosure, in which one provisional account hardens too early and narrows the space in which alternative meanings can be articulated and heard.

“The resulting risk is not only decision error.It is interpretive foreclosure, in which one provisional account hardens too early and narrows the space in which alternative meanings can be articulated and heard.”

• AssumptionRests on an assumption: Preserving human participation in the interpretation of their own lives is a good that institutions should protect.

Classified as probably not a substantial claim.

Ricoeur's account of narrative identity holds that persons are not only objects described from outside; they participate in narrating continuity, action and responsibility across time.

“Ricoeur's account of narrative identity is relevant because persons are not only objects described from outside; they participate in narrating continuity, action and responsibility across time (Ricoeur, 1992).”

• cites Ricoeur, 1992Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Generative AI can reduce completion time and improve average quality in professional writing tasks.

“Experimental evidence shows that generative AI can reduce completion time and improve average quality in professional writing tasks (Noy & Zhang, 2023).”

• cites Noy & Zhang, 2023 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Oziev (2026) develops hermeneutic sovereignty as a constitutional court's capacity to manage legal doubt and produce authoritative legal certainty.

“Oziev (2026) develops hermeneutic sovereignty as a constitutional court's capacity to manage legal doubt and produce authoritative legal certainty.”

• cites Oziev (2026)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Human-centred guidance in education emphasises ethical validation, human agency and appropriate institutional capacity.

“Human-centred guidance in education likewise emphasises ethical validation, human agency and appropriate institutional capacity (Miao & Holmes, 2023).”

• cites Miao & Holmes, 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

In creative tasks, generative AI can increase individual performance while making outputs more similar at the collective level.

“In creative tasks, it can also increase individual performance while making outputs more similar at the collective level (Doshi & Hauser, 2024).”

• cites Doshi & Hauser, 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

A generated summary is not simply a shorter version of an original; it selects salience, orders causes, establishes tone, chooses categories and suppresses some ambiguities so that others can be foregrounded.

“A generated summary is never simply a shorter version of an original.It selects salience, orders causes, establishes tone, chooses categories and suppresses some ambiguities so that others can be foregrounded.”

Classified as probably not a substantial claim.

Pohlhaus (2012) shows how dominant knowers can refuse interpretive resources developed from marginalised experience.

“Pohlhaus (2012) shows how dominant knowers can refuse interpretive resources developed from marginalised experience.”

• cites Pohlhaus (2012) sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Its requirements rise with stakes, persistence, asymmetry of power and difficulty of reversal.

“Its requirements rise with stakes, persistence, asymmetry of power and difficulty of reversal.”

May combine two claims.

Interpretation is shaped by prior understandings and becomes productive through dialogue.

“Gadamer (2004) emphasises that interpretation is shaped by prior understandings and becomes productive through dialogue.”

• cites Gadamer (2004)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Generative AI can increase individual creative performance while reducing collective diversity.

“it can increase individual creative performance while reducing collective diversity (Doshi & Hauser, 2024).”

• cites Doshi & Hauser, 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

Dotson (2011) analyses forms of silencing that occur when audiences fail to meet speakers' epistemic vulnerabilities.

“Dotson (2011) analyses forms of silencing that occur when audiences fail to meet speakers' epistemic vulnerabilities.”

• cites Dotson (2011) sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

The conditions under which meaning appears are themselves part of what must be examined.

“It means that the conditions under which meaning appears are themselves part of what must be examined.”

Probably not a substantial claim: may not pass S2 (checkable).

The more persuasive the prose, the easier it becomes to mistake an interpretation for the underlying material itself.

“The more persuasive the prose, the easier it becomes to mistake an interpretation for the underlying material itself.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Sovereignty is relational, not atomistic.

“Second, it is relational, not atomistic.”

• AssumptionRests on an assumption: Meaning is socially formed.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Persons cannot reasonably demand control over every external interpretation of them.

“Persons cannot reasonably demand control over every external interpretation of them.”

• AssumptionRests on an assumption: Social life necessarily involves interpretation by others.

Classified as probably not a substantial claim.

The framework is designed for empirical challenge: its dimensions generate observable questions about workflow design, record persistence, contestation and revision.

“Finally, the framework is designed for empirical challenge: its dimensions generate observable questions about workflow design, record persistence, contestation and revision.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Medina (2013) emphasises epistemic resistance and the importance of friction among differently situated perspectives.

“Medina (2013) emphasises epistemic resistance and the importance of friction among differently situated perspectives.”

• cites Medina (2013) sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

Sovereignty concerns standing as well as capacity.

“Third, it concerns standing as well as capacity.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Sovereignty is situated, not absolute.

“First, sovereignty is situated, not absolute.”

May combine two claims.

A large language model can propose the issue, select the facts, generate the causal link, assign the tone and recommend the conclusion.

“It can propose the issue, select the facts, generate the causal link, assign the tone and recommend the conclusion.”

Classified as probably not a substantial claim. May combine two claims.

Cognitive offloading can reduce immediate mental demands while also redistributing the work required for memory, evaluation and control.

“Cognitive offloading can reduce immediate mental demands while also redistributing the work required for memory, evaluation and control (Risko & Gilbert, 2016).”

• cites Risko & Gilbert, 2016 sourceNot checked — no full text of the source

Classified as probably not a substantial claim. May combine two claims.

The machine narrative arrives early, polished and aligned with institutional categories; the person's reply arrives later, perhaps emotionally, incompletely or outside the preferred template.

“The machine narrative arrives early, polished and aligned with institutional categories; the person's reply arrives later, perhaps emotionally, incompletely or outside the preferred template.”

Probably not a substantial claim: may not pass S2 (checkable).

Smoothness can obscure the fact that a synthesis is an achievement of framing rather than a transparent view onto reality.

“Smoothness can be useful, but it can also obscure the fact that a synthesis is an achievement of framing rather than a transparent view onto reality.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Reliance on automation depends on trust calibration, system characteristics and context.

“Reliance on automation depends on trust calibration, system characteristics and context (Lee & See, 2004).”

• cites Lee & See, 2004 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Generative AI can increase productivity and apparent quality in professional tasks.

“Generative AI can increase productivity and apparent quality in professional tasks (Noy & Zhang, 2023)”

• cites Noy & Zhang, 2023 sourceNot checked — no full text of the source

May combine two claims.

Fluent assistance can narrow human exploration while improving surface-level performance.

“These findings do not prove interpretive foreclosure, but they support the plausibility of mechanisms through which fluent assistance can narrow human exploration while improving surface-level performance.”

Classified as probably not a substantial claim.

This creates an asymmetry of sequence and format.

“This creates an asymmetry of sequence and format.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

A person may have the intellectual ability to contest an interpretation but lack an authorised channel through which the contest can matter.

“A person may have the intellectual ability to contest an interpretation but lack an authorised channel through which the contest can matter.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Claims about existing theory are separated from the paper's new constructs.

“Claims about existing theory are separated from the paper's new constructs.”

Probably not a substantial claim: may not pass S2 (checkable).

If a model cannot bear responsibility for understanding, then the human and institutional processes that adopt its representations become more important.

“If a model cannot bear responsibility for understanding, then the human and institutional processes that adopt its representations become more important.”

May combine two claims.

AI tools can produce illusions of understanding and scientific monocultures, where apparent productivity coexists with reduced epistemic diversity.

“In science, Messeri and Crockett (2024) warn that AI tools can produce illusions of understanding and scientific monocultures, where apparent productivity coexists with reduced epistemic diversity.”

• cites Messeri and Crockett (2024) sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

When such outputs are embedded in routine workflows, the system becomes interpretive infrastructure: an environment through which meaning is produced and circulated.

“When such outputs are embedded in routine workflows, the system becomes interpretive infrastructure: an environment through which meaning is produced and circulated.”

Classified as probably not a substantial claim.

Empirical studies are used only for the phenomena they establish, so productivity gains, output similarity, automation reliance and illusions of understanding are evidence of relevant tendencies rather than direct proof of institutional injustice.

“Empirical studies are used only for the phenomena they establish, so productivity gains, output similarity, automation reliance and illusions of understanding are evidence of relevant tendencies rather than direct proof of institutional injustice.”

Probably not a substantial claim: may not pass S4 (based).

An appeal channel may formally exist while the person lacks access to the source material or sufficient explanation to use it.

“Conversely, an appeal channel may formally exist while the person lacks access to the source material or sufficient explanation to use it.”

Probably not a substantial claim: may not pass S2 (checkable).

Not every interpretation is equally valid.

“This does not imply that every interpretation is equally valid.”

Probably not a substantial claim: may not pass S4 (based). May combine two claims.

Once an AI-generated summary enters a record, subsequent systems may retrieve it, compress it again, use it as context for another generation, or present it to future decisionmakers.

“Once an AI-generated summary enters a record, subsequent systems may retrieve it, compress it again, use it as context for another generation, or present it to future decisionmakers.”

Classified as probably not a substantial claim.

The paper offers a theory and governance framework, not causal estimates.

“The principal limitation is therefore intentional: the paper offers a theory and governance framework, not causal estimates.”

Classified as probably not a substantial claim.

A case file containing uncertainty, contradiction and incomplete testimony can be transformed into a smooth paragraph.

“A case file containing uncertainty, contradiction and incomplete testimony can be transformed into a smooth paragraph.”

Probably not a substantial claim: may not pass S2 (checkable).

This is a distinct form of power: not the power to make one decision, but the power to establish the story from which later decisions begin.

“This is a distinct form of power: not the power to make one decision, but the power to establish the story from which later decisions begin.”

Probably not a substantial claim: may not pass S2 (checkable).

Personal identity is narrated through action, memory, character, promise and revision across time.

“It is narrated through action, memory, character, promise and revision across time (Ricoeur, 1992).”

• cites Ricoeur, 1992Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S4 (based).

A phrase such as 'resistant to change' can migrate from one AI-assisted review into succession planning, development recommendations and later prompts.

“A phrase such as”

Probably not a substantial claim: may not pass S2 (checkable).

Explanation does not create standing if the affected person cannot append a counteraccount.

“Explanation does not create standing if the affected person cannot append a counteraccount.”

Classified as probably not a substantial claim.

The Interpretive Foreclosure Test complements rather than replaces existing AI risk management.

“The test complements rather than replaces existing AI risk management.”

Probably not a substantial claim: may not pass S2 (checkable).

Without a propagation-aware correction process, an initially weak interpretation can become organisational memory.

“Without a propagation-aware correction process, an initially weak interpretation can become organisational memory.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Correction is often local while propagation is systemic.

“This mechanism is particularly important because correction is often local while propagation is systemic.”

Classified as probably not a substantial claim. May combine two claims.

Deliberate plural prompting and temporal separation introduce productive friction.

“These practices introduce productive friction.”

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

The human can attribute the wording to the system; the system cannot answer for the judgment.

“The human can attribute the wording to the system; the system cannot answer for the judgment.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

A human reviewer may retain formal authority while narrative authorship, contextual integrity and plurality have already been lost.

“A human reviewer may retain formal authority while narrative authorship, contextual integrity and plurality have already been lost.”

Probably not a substantial claim: may not pass S4 (based).

When AI synthesises emails, project data or prior feedback, it can transform scattered observations into an authoritative organisational identity.

“When AI synthesises emails, project data or prior feedback, it can transform scattered observations into an authoritative organisational identity.”

• AssumptionRests on an assumption: Generative AI is particularly attractive in performance management because managers face large documentation burdens.

Probably not a substantial claim: may not pass S3 (load-bearing).

The same logic responds to the risk of monoculture identified by Messeri and Crockett.

“The same logic responds to the risk of monoculture identified by Messeri and Crockett (2024).”

• cites Messeri and Crockett, 2024 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

Hermeneutic sovereignty does not require employees to control performance judgments.

“Hermeneutic sovereignty does not require employees to control performance judgments.”

Probably not a substantial claim: may not pass S4 (based).

Performance reviews narrate contribution, potential, attitude and trajectory.

“They narrate contribution, potential, attitude and trajectory.”

May combine two claims.

AI assistance can improve individual outputs while reducing collective diversity.

“Doshi and Hauser (2024) find that AI assistance can improve individual outputs while reducing collective diversity.”

• cites Doshi and Hauser (2024)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

Sakabe's account of interpretive sovereignty in creative authorship and Doshi and Hauser's empirical findings both point toward a more nuanced relationship.

“Sakabe's (2026) account of interpretive sovereignty in creative authorship and Doshi and Hauser's (2024) empirical findings both point toward a more nuanced relationship.”

• cites Sakabe, 2026 sourceNot checked — no full text of the source
• cites Doshi and Hauser, 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Responsibility is diffused between user, model, vendor and workflow.

“Responsibility is diffused between user, model, vendor and workflow.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Appeals address outcomes while the underlying narrative remains fixed.

“Appeals address outcomes while the underlying narrative remains fixed.”

Probably not a substantial claim: may not pass S4 (based).

Hermeneutic sovereignty can be evaluated through six dimensions.

“Hermeneutic sovereignty can be evaluated through six dimensions.”

Probably not a substantial claim: may not pass S2 (checkable).

Recursive fixation transforms a temporal convenience into a governance problem.

“Recursive fixation transforms a temporal convenience into a governance problem.”

Probably not a substantial claim: may not pass S2 (checkable).

Deliberate plural prompting and temporal separation preserve the user's role as an originator and evaluator rather than turning review into acceptance of the first fluent proposal.

“They preserve the user's role as an originator and evaluator rather than turning review into acceptance of the first fluent proposal.”

May combine two claims.

When a common model, prompt library or organisational template structures many cases, the institution may become more linguistically consistent while becoming less hermeneutically plural.

“When a common model, prompt library or organisational template structures many cases, the institution may become more linguistically consistent while becoming less hermeneutically plural.”

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

AI supports sovereignty when it expands the person's repertoire while keeping alternatives visible and preserving the person's capacity to revise or refuse.

“AI supports sovereignty when it expands the person's repertoire while keeping alternatives visible and preserving the person's capacity to revise or refuse.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Machine framing becomes the default starting narrative.

“Machine framing becomes the default starting narrative.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based). May combine two claims.

Early labels persist and become self-confirming.

“Early labels persist and become self-confirming.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

One fluent synthesis suppresses alternatives too early.

“One fluent synthesis suppresses alternatives too early.”

Classified as probably not a substantial claim.

All three elements of hermeneutic sovereignty-standing, capacity and uptake-are necessary.

“All three are necessary.”

Probably not a substantial claim: may not pass S3 (load-bearing).

UNESCO guidance correctly emphasises human-centred use, capacity development and pedagogical validation.

“UNESCO guidance correctly emphasises human-centred use, capacity development and pedagogical validation (Miao & Holmes, 2023).”

• cites Miao & Holmes, 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

The AI does not possess institutional legitimacy, yet its fluency can mask the point at which discretionary choices entered the process.

“The AI does not possess institutional legitimacy, yet its fluency can mask the point at which discretionary choices entered the process.”

May combine two claims.

Generative AI can increase access to interpretive resources and counter some forms of epistemic disadvantage.

“Generative AI can therefore increase access to interpretive resources and counter some forms of epistemic disadvantage.”

May combine two claims.

AI-supported science can encourage illusions of understanding, where researchers feel they comprehend more than they actually do, while the scientific community becomes vulnerable to monocultures.

“Messeri and Crockett (2024) show how AI-supported science can encourage illusions of understanding, where researchers feel that they comprehend more than they actually do, while the scientific community becomes vulnerable to monocultures.”

• cites Messeri and Crockett (2024) sourceNot checked — no full text of the source

Classified as probably not a substantial claim.

A formal right to contest can be hollow if sources are unavailable or the generated narrative has already propagated.

“A formal right to contest can also be hollow if sources are unavailable or the generated narrative has already propagated.”

Probably not a substantial claim: may not pass S2 (checkable).

A comment box offers little uptake if no reviewer must address the correction.

“A comment box offers little uptake if no reviewer must address the correction.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Recursive fixation connects hermeneutic sovereignty to data lifecycle management.

“It also connects hermeneutic sovereignty to data lifecycle management.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Performance reviews are not neutral compilations of facts.

“Yet performance reviews are not neutral compilations of facts.”

May combine two claims.

AI does not create interpretation or injustice for the first time; it changes their scale, speed, standardisation and persistence.

“AI does not create interpretation or injustice for the first time.It changes their scale, speed, standardisation and persistence.”

• AssumptionRests on an assumption: Generative systems can produce polished summaries instantly, propagate them across organisational systems and normalise a shared style of reasoning, while responsibility becomes easier to diffuse.

Probably not a substantial claim: may not pass S3 (load-bearing).

Research on automation shows that trust and reliance are shaped by context, perceived competence and system presentation.

“Research on automation shows that trust and reliance are shaped by context, perceived competence and system presentation (Lee & See, 2004).”

• cites Lee & See, 2004 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Rich circumstances are compressed into administratively convenient categories.

“Rich circumstances are compressed into administratively convenient categories.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

The alternative interpretation need not be correct.

“The alternative interpretation need not be correct.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

A workflow can use substantial AI assistance while preserving sovereignty if human accounts are elicited first, uncertainty is maintained, sources remain inspectable, alternatives can be expressed, corrections propagate and accountable humans own the final interpretation.

“Conversely, a workflow can use substantial AI assistance while preserving sovereignty if human accounts are elicited first, uncertainty is maintained, sources remain inspectable, alternatives can be expressed, corrections propagate and accountable humans own the final interpretation.”

Probably not a substantial claim: may not pass S2 (checkable).

AI undermines sovereignty when the generated frame becomes difficult to escape because of sequence, authority or persistence.

“It undermines sovereignty when the generated frame becomes difficult to escape because of sequence, authority or persistence.”

Classified as probably not a substantial claim.

A positive answer to one question on the Interpretive Foreclosure Test is not automatically unacceptable.

“A positive answer to one question is not automatically unacceptable.”

Classified as probably not a substantial claim. May combine two claims.

Cognitive offloading research shows that people strategically externalise cognitive work, sometimes based on imperfect metacognitive judgments.

“Cognitive offloading research likewise shows that people strategically externalise cognitive work, sometimes based on imperfect metacognitive judgments (Risko & Gilbert, 2016).”

• cites Risko & Gilbert, 2016 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

The six dimensions function as normative and diagnostic dimensions that can be translated into domain-specific indicators.

“They function as normative and diagnostic dimensions that can be translated into domain-specific indicators.”

Probably not a substantial claim: may not pass S2 (checkable).

Cognitive sovereignty extends from knowing and checking toward self-interpretation in institutional contexts.

“This extends cognitive sovereignty from knowing and checking toward self-interpretation in institutional contexts (Tan, 2026b).”

• cites Tan, 2026bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable).

The configuration of all four conditions creates a pathway from convenience to interpretive domination.

“That configuration creates a pathway from convenience to interpretive domination.”

Classified as probably not a substantial claim.

Risk becomes especially serious when all four conditions of the Interpretive Foreclosure Test align: the machine speaks first, compresses complexity, acquires institutional authority and persists over time.

“Risk becomes especially serious when all four conditions align: the machine speaks first, compresses complexity, acquires institutional authority and persists over time.”

May combine two claims.

Institutional interpretation can exhibit an analogous tension: standardised generated language may raise average clarity and consistency, yet also reduce the variety of ways in which ambiguous situations are described.

“Institutional interpretation can exhibit an analogous tension.Standardised generated language may raise average clarity and consistency, yet also reduce the variety of ways in which ambiguous situations are described.”

Probably not a substantial claim: may not pass S2 (checkable).

The concern arises when the system becomes the default source of interpretation rather than an interlocutor within a human-led process.

“The concern arises when the system becomes the default source of interpretation rather than an interlocutor within a human-led process.”

Probably not a substantial claim: may not pass S2 (checkable).

This infrastructural role is easy to overlook because the interface often presents generation as assistance.

“This infrastructural role is easy to overlook because the interface often presents generation as assistance.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Foreclosure need not be deliberate and can emerge from convenience, incentives and the rhetorical force of fluent synthesis.

“Foreclosure need not be deliberate and can emerge from convenience, incentives and the rhetorical force of fluent synthesis.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Both approaches point toward a broader question of human control under AI mediation.

“Both approaches point toward a broader question of human control under AI mediation.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

The distinctive risk of generative AI is the speed, scale and linguistic completeness with which compression can occur.

“The distinctive risk of generative AI is the speed, scale and linguistic completeness with which compression can occur.”

• AssumptionRests on an assumption: All institutions compress information.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

A technically safe model can still participate in an unfair meaning chain if the surrounding workflow gives its interpretations premature or irreversible authority.

“A technically safe model can still participate in an unfair meaning chain if the surrounding workflow gives its interpretations premature or irreversible authority.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Some friction forces source inspection, comparison and reason-giving.

“Some friction is valuable because it forces source inspection, comparison and reason-giving.”

Probably not a substantial claim: may not pass S4 (based).

Four mechanisms explain how hermeneutic sovereignty can be weakened: interpretive pre-emption, hermeneutic compression, authority laundering and recursive fixation.

“Four mechanisms explain how this standing can be weakened: interpretive pre-emption, hermeneutic compression, authority laundering and recursive fixation.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

These requirements also clarify the limits of general AI ethics principles.

“These requirements also clarify the limits of general AI ethics principles.”

Probably not a substantial claim: may not pass S2 (checkable).

Requiring institutions to know who adopted an account, what evidence was relied upon, what material uncertainty remained, and whether the affected person's account was considered re-establishes responsibility at the point where machine prose becomes institutional action.

“This re-establishes responsibility at the point where machine prose becomes institutional action.”

Probably not a substantial claim: may not pass S4 (based).

A system can be documented, privacy-preserving and technically reliable while still narrowing the interpretive standing of those it represents.

“A system can be documented, privacy-preserving and technically reliable while still narrowing the interpretive standing of those it represents.”

• AssumptionRests on an assumption: Overdocumentation can itself burden and surveil.

Probably not a substantial claim: may not pass S2 (checkable).

When machine-generated interpretations become the starting point for evaluation, administration or professional judgment, a new humanities problem emerges: the person may remain formally present while losing practical standing in the construction of meaning.

“When machine-generated interpretations become the starting point for evaluation, administration or professional judgment, a new humanities problem emerges: the person may remain formally present while losing practical standing in the construction of meaning.”

Probably not a substantial claim: may not pass S2 (checkable).

Generative AI changes the governance of interpretation.

“Generative AI changes the governance of interpretation.”

Probably not a substantial claim: may not pass S2 (checkable).

Together the four mechanisms describe AI-mediated interpretive foreclosure, a process through which provisional machine framings can become authoritative before adequate human interpretation has occurred.

“Together they describe AI-mediated interpretive foreclosure, a process through which provisional machine framings can become authoritative before adequate human interpretation has occurred.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Tan (2026b) develops cognitive sovereignty as the capacity to originate, evaluate, contest, retain and responsibly own knowledge claims even when computational assistance is used.

“Tan (2026b) develops cognitive sovereignty as the capacity to originate, evaluate, contest, retain and responsibly own knowledge claims even when computational assistance is used.”

• cites Tan (2026b)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

If revision does not follow derived summaries, dashboards and downstream systems, the formal right to correction is defeated by recursive fixation.

“Otherwise, the formal right to correction is defeated by recursive fixation.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

The significance of generative AI lies not only in faster drafting, better search or automated recommendation, but in its growing ability to supply the language through which institutions understand people.

“Its significance lies not only in faster drafting, better search or automated recommendation, but in its growing ability to supply the language through which institutions understand people.”