CoChat
Sign Up
PreprintCited by 7Innovative Education and Learning PracticesHuman Resource Development and Performance EvaluationAI and HR Technologies

The Tragedy of the Cognitive Commons: How AI Could Disrupt the Regeneration of Professional Expertise

Open access
Lovett, Nolan·2026·Human Resource Development Review
DOI 10.1177/15344843261470602arXiv 2607.29380
Open this paper in CoChat — free accountOpen PDF

Abstract

Artificial intelligence is reshaping cognitive work, but Human Resource Development scholarship has treated this transformation as an organizational training challenge, leaving the collective regeneration of professional expertise unexamined. This conceptual paper introduces the Cognitive Commons framework, integrating commons theory, HRD scholarship, and distributed cognition to explain how rational AI adoption decisions can deplete the shared expertise pool professions require for renewal. The framework distinguishes Internalized Mastery (deep domain knowledge from sustained practice) from Distributed Mastery (orchestrating human-AI systems), and develops the Validation Tether: effective AI oversight depends on the expertise AI adoption may undermine. Early labor market and clinical evidence suggests possible disruption to expertise-regeneration pathways in highly AI-exposed sectors, though adoption is recent and the strongest signals come from leading sectors rather than all professions. Five factors determine occupational vulnerability, and governance arrangements may form across organizational, professional-association, and policy levels. The paper reframes expertise development as collective stewardship rather than organizational optimization, with implications for HRD theory and workforce policy.

AI snapshot

Written from the paper’s full text.

TL;DR

This conceptual paper introduces the "Cognitive Commons" framework, arguing that rational organizational AI adoption can deplete the profession-level pool of deep human expertise on which all organizations depend, because AI eliminates the entry-level work through which expertise has historically regenerated. It distinguishes Internalized Mastery from Distributed Mastery and proposes the "Validation Tether": effective oversight of AI depends on the very deep domain expertise that AI adoption erodes.

Key findings

  • The framework defines the Cognitive Commons as the collective pool of deep human expertise within a profession, arguing it meets three commons conditions: collective dependence, non-excludability, and degradability through overexploitation of the regeneration mechanism. Depletion is framed as a structural prediction from the empirically established dissociation between assisted performance and independent capability, "not an observed outcome."
  • Cited labor-market evidence: Brynjolfsson et al. (2025), using payroll data covering over 25 million U.S. workers (analysis sample of 3.5–5 million per month), found workers ages 22–25 in the most AI-exposed occupations experienced a 16% relative employment decline between October 2022 and September 2025, while workers aged 35–49 in those occupations grew by over 8%; Hampole et al. (2025) using 58 million LinkedIn profiles is cited as reinforcing the cohort-specific pattern.
  • Cited validation evidence: Vicente and Matute (2023) found 80.7% of participants detected errors in biased AI recommendations yet continued following erroneous advice; Niederhoffer et al. (2025) found 40% of surveyed full-time employees received substantively flawed AI-generated content in the past month, spending nearly two hours on average per instance; Benzing et al. (2025) found 60% of employees reported confidence enough in AI output that they do not routinely check its accuracy.
  • Five factors are proposed to determine occupational vulnerability: task substitutability, regulatory intensity, safety criticality, professional association strength, and work modularization. Software engineering, financial analysis, and legal research are placed in higher-vulnerability categories; medicine and engineering in lower ones due to regulatory and safety counterpressures.

Methods

  • Conceptual paper integrating commons theory, HRD and adult education scholarship, distributed cognition, and systems thinking, following Jaakkola's (2020) framework for integrating fragmented literatures; construct development is positioned per Rocco et al. (2022) as preceding operationalization and empirical testing.
  • No primary data are collected. The argument draws on secondary evidence from labor economics (US, UK, German labor markets), clinical medicine, and experimental psychology, and formalizes relationships in a six-node causal chain of commons depletion (Figure 1) plus construct definitions (Table 1) and a proposed indicator research agenda (Table 2).
  • The paper distinguishes three evidential levels for its claims: empirically grounded claims, theoretically derived claims with corroborating indirect evidence, and extrapolatory predictions requiring direct empirical test.

Limitations

  • Stated: the data span less than three years of widespread generative AI adoption, representing early-stage rather than mature dynamics; patterns concentrate in the most AI-exposed occupations and do not characterize professional employment broadly; the evidence does not demonstrate widespread validation failures or profession-wide expertise collapse. The paper also states the profession-level depletion is a structural prediction, not an observed outcome.
  • Stated boundary conditions: the framework assumes a Western cultural model of professional formation (entry-level employment in hierarchical organizations, formal credentialing) and assumes AI systems continue to require human oversight; it excludes professions facing wholesale disintermediation.
  • Evident from the text: the framework's core claims rest on secondary sources and no original empirical test is conducted; the paper itself notes that the Brynjolfsson et al. data capture only direct position elimination and "cannot detect Mechanism 2" (augmentation without internalization), and that countervailing evidence (e.g., Humlum & Vestergaard's null effects in Denmark, Manning & Aguirre's adaptive capacity findings) establishes the depletion mechanism is not universal.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1177/15344843261470602

Cites

68 of 68 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

35 of 105 cited sources checked · 16 supported · 1 flagged · accuracy not yet measured

  • Premisecitation

    The dissociation between AI-assisted performance and independent capability is empirically established by studies that measure unaided performance after AI exposure.

    “The dissociation between assisted performance and independent capability is empirically established, directly measured wherever studies test unaided performance after AI exposure (Budzyń et al., 2025;Macnamara et al., 2024).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Budzyń et al., 2025 source
    • cites Macnamara et al., 2024 source

      In “Potential AI-induced skill development hindrance” (page unavailable), the comparison of AI-assisted performance with later unaided performance is explicitly hypothetical: the authors say “suppose” AI is withheld and “we might expect” worse performance. In “The use of automation and artificial intelligence” (page unavailable), they state that, to their awareness, no research had examined AI assistants’ effects on users’ skill. This paper therefore does not present the dissociation as empirically established by such tests. (LLM, conf=0.88)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
  • Premisecitation

    In the most AI-exposed sectors, organizations are already eliminating entry-level positions that historically provided the foundation for expertise development.

    “In the most AI-exposed sectors, organizations are already eliminating the entry-level positions that have historically provided the foundation for expertise development (Brynjolfsson et al., 2025;Hampole et al., 2025), and these disruptions continue to be interpreted as temporary workforce adjustments rather than as signals of regeneration failure.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Brynjolfsson et al., 2025

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

    • cites Hampole et al., 2025 source

      The Abstract and §3.5 report AI-related declines in employment share for some occupations, including business, financial, and engineering roles. They do not establish that organizations are eliminating entry-level positions or that those positions historically provided a foundation for expertise development; searches for entry-level hiring and career development did not find that link in our copy. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
  • Interpretationcitation

    The erosion of the Cognitive Commons becomes measurable through cohort-level labor market analysis showing systematic elimination of early-career positions.

    “The erosion becomes measurable through cohort-level labor market analysis showing systematic elimination of early-career positions (Brynjolfsson et al., 2025;Hampole et al., 2025), but organizations and policymakers commonly interpret these patterns as temporary labor market adjustment”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Brynjolfsson et al., 2025

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

    • cites Hampole et al., 2025 source

      The paper’s Abstract reports reduced demand for AI-exposed tasks and some occupational employment-share declines. Its analysis compares tasks and occupations within firms, however, not early-career cohorts; it does not show systematic elimination of early-career positions. The Abstract also describes overall employment effects as modest. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Premisecitation

    Collaborative AI workflows requiring active clinical engagement have improved diagnostic accuracy rather than degrading it.

    “collaborative AI workflows requiring active clinical engagement have improved diagnostic accuracy rather than degrading it (Everett et al., 2025).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Everett et al., 2025 source
  • Interpretationcitation

    Deep expertise develops through scaffolded engagement with progressively complex problems, in which practitioners struggle with difficulty, monitor and revise their own reasoning, and repeatedly correct flawed assumptions against evidence.

    “Deep expertise develops through scaffolded engagement with progressively complex problems, in which practitioners struggle with difficulty, monitor and revise their own reasoning, and repeatedly correct flawed assumptions against evidence (Ericsson, 2006;Mezirow, 1991;Rind, 2022).”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Ericsson, 2006 source

      Ericsson says that developing expert performance requires deliberate efforts to change aspects of performance (p. 685), instruction by teachers (p. 692), and increasingly complex mental representations that give performers greater control (Figure 38.1, p. 687). The available passages do not establish the more specific account of progressively complex problems, monitoring and revising one’s reasoning, or repeatedly correcting flawed assumptions against evidence. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites Mezirow, 1991 source
    • cites Rind, 2022 source
  • Classified as probably not a substantial claim. 55
  • May combine two claims. 57
  • The quote only partly states this claim. 1

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

    Premisereasoning

    The early empirical evidence reviewed draws on US, UK, and German labor markets, reflecting the current state of the research base rather than a theoretical limitation.

    “The early empirical evidence reviewed in this paper draws on US (Brynjolfsson et al., 2025;Hampole et al., 2025), UK (Klein Teeselink, 2025), and German (Engberg et al., 2025) labor markets, reflecting the current state of the research base rather than a theoretical limitation.”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Brynjolfsson et al., 2025

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

    • cites Hampole et al., 2025 source
    • cites Klein Teeselink, 2025 source

  • May combine two claims.

    Premisecitation

    Expertise is built through informal learning at work, legitimate peripheral participation, transformative learning, and prolonged deliberate practice.

    “expertise develops through informal workplace learning (Eraut, 2004;Blume et al., 2010), legitimate peripheral participation (Lave & Wenger, 1991), transformative learning (Mezirow, 1991), and sustained deliberate practice (Ericsson, 2006).”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites Eraut, 2004 source
    • cites Lave & Wenger, 1991 source
    • cites Ericsson, 2006 source

  • May combine two claims.

    Premisecitation

    Workplace learning dominates formal training in producing professional expertise, operating through legitimate peripheral participation.

    “This workplace learning dominates formal training in producing professional expertise (Eraut, 2004;McCall et al., 1988), operating through what Lave and Wenger (1991) term legitimate peripheral participation.”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Eraut, 2004 source
    • cites McCall et al., 1988

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

    • cites Lave and Wenger, 1991 source

  • May combine two claims.

    Premisecitation

    Aggregate labor market evidence is mixed, with studies documenting adaptive capacity, considerable heterogeneity in retrainability across occupations and labor market conditions, and null effects on earnings and recorded hours among workers in Denmark during the first two years of generative AI adoption.

    “More broadly, the aggregate labor market evidence is mixed, with studies documenting adaptive capacity (Manning & Aguirre, 2026), considerable heterogeneity in retrainability across occupations and labor market conditions (Hyman et al., 2025), and null effects on earnings and recorded hours among workers in Denmark during the first two years of generative AI adoption (Humlum & Vestergaard, 2025).”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites Manning & Aguirre, 2026 source
    • cites Hyman et al., 2025 source
    • cites Humlum & Vestergaard, 2025

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

  • May combine two claims.

    Interpretationcitation

    Practitioners who treat AI outputs as authoritative tend to accept fluent answers without scrutiny, while those with more sophisticated epistemic orientations verify, cross-reference, and question before accepting.

    “Practitioners who treat AI outputs as authoritative tend to accept fluent answers without scrutiny, while those with more sophisticated epistemic orientations verify, cross-reference, and question before accepting (Rind et al., 2026).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Rind et al., 2026 source

      The Findings describe students who treated ChatGPT as an infallible authority and rarely verified its outputs, in contrast to critical evaluators who questioned and cross-referenced them. The claim extends this pattern to practitioners generally, while the paper’s findings concern students. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    80.7% of participants detected errors in the AI recommendations, showing functional surface validation.

    “Critically, 80.7% of participants detected errors in the AI recommendations, demonstrating functional surface validation.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vicente and Matute (2023) source

  • May combine two claims.

    Interpretationcitation

    When AI assistance preserves the practitioner's own attempt and requires them to evaluate, question, and justify, the new cognitive work can be developmental; when it supplies finished outputs that bypass that effort, augmentation weakens Internalized Mastery formation.

    “Where AI assistance preserves the practitioner's own attempt and requires them to evaluate, question, and justify, the new cognitive work can itself be developmental; where it supplies finished outputs that bypass that effort, augmentation attenuates Internalized Mastery formation (Rind, 2026).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Rind, 2026 source

      Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    HRD education now prepares professionals for AI-integrated workplaces through generative AI applications in instructional design, responsible AI training frameworks, and technology competence development.

    “HRD education now prepares professionals for AI-integrated workplaces through generative AI applications in instructional design (Chai et al., 2025), responsible AI training frameworks (Chen, 2024), and technology competence development (Ekuma, 2024).”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Chai et al., 2025 source
    • cites Chen, 2024 source
    • cites Ekuma, 2024 source

  • May combine two claims.

    Findingcitation

    Skill carryover from AI-assisted ideation tasks is limited, with the effect strongest on tasks requiring domain judgment.

    “Kumar et al. (2025) andBangerl et al. (2025) found limited skill carryover from AI-assisted ideation tasks, with the effect most pronounced on tasks requiring domain judgment.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Kumar et al. (2025) source

      The Introduction reports that LLM assistance did not improve performance on subsequent unassisted creative tasks, supporting limited skill carryover. Section 4.2.1 reports poorer unassisted performance for the LLM Guidance group on a word-association task, but the paper does not establish that the effect is strongest on tasks requiring domain judgment. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites Bangerl et al. (2025) source

  • May combine two claims.

    Premisecitation

    Methodological templates already exist, including task-level AI exposure measures from resume data, experimental paradigms distinguishing task completion from substantive validation capability, and analysis of cognitive skill erosion.

    “Methodological templates already exist, including Hampole et al.'s (2025) task-level AI exposure measures from resume data, Vicente and Matute's (2023) experimental paradigms distinguishing task completion from substantive validation capability, and Macnamara et al.'s (2024) analysis of cognitive skill erosion.”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites Hampole et al. (2025) source
    • cites Vicente and Matute (2023) source
    • cites Macnamara et al. (2024) source

  • May combine two claims.

    Findingcitation

    Workers aged 22 to 25 in AI-exposed occupations have experienced employment declines while experienced workers' employment has remained stable.

    “A convergence of empirical evidence indicates that the regeneration mechanism faces disruption in affected sectors: workers ages 22 to 25 in AI-exposed occupations have experienced employment declines while experienced workers remain stable”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Brynjolfsson et al. (2025)

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

    • cites Hampole et al. (2025) source

      Section 3.3, “AI Exposure and Occupational Labor Demand” (page unavailable), reports that greater mean AI task exposure predicts employment declines in occupation–firm pairs. The cited passages do not report results for workers aged 22–25 or compare their employment with that of experienced workers; searches for those age and experience terms found no such comparison. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingcitation

    40% of surveyed full-time employees received substantively flawed AI-generated content in the past month, and recipients spent nearly two hours on average addressing each instance.

    “Niederhoffer et al. (2025) found that 40% of surveyed full-time employees received substantively flawed AI-generated content in the past month, with recipients spending nearly two hours on average addressing each instance”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Niederhoffer et al. (2025)

      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).

    Interpretationreasoning

    The appropriate unit of analysis for commons dynamics is the occupation rather than the national economy.

    “The appropriate unit of analysis for commons dynamics is the occupation rather than the national economy.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    The framework's structural conditions are not specific to any national context; the commons dynamic operates wherever AI substitutes for developmental cognitive labor within professions characterized by permeable labor markets, collective dependence on shared expertise, and organizational rather than educational developmental pathways.

    “The framework's structural conditions are not specific to any national context.The commons dynamic operates wherever AI substitutes for developmental cognitive labor within professions characterized by permeable labor markets, collective dependence on shared expertise, and organizational rather than educational developmental pathways.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Rational calculation favors appropriation over contribution, and organizations investing in developmental infrastructure face competitive disadvantage against those hiring already-trained experts.

    “The rational calculation favors appropriation over contribution.Organizations that invest in developmental infrastructure, whether preserving traditional pipelines or building new ones, face competitive disadvantage against those that rely on hiring already-trained experts from the shrinking pool.”
    Quote in the paperSubstantial claim

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

    Premisenone given

    The cultural model of professional formation in which expertise develops through entry-level employment in hierarchical organizations and is governed through formal credentialing institutions dominates Western professional sectors and the evidence base reviewed, but does not exhaust the ways human expertise has been or can be transmitted.

    “This model dominates Western professional sectors and the evidence base reviewed here, but it does not exhaust the ways human expertise has been or can be transmitted.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The Cognitive Commons is collectively dependent: organizations rely on hiring experts from a shared talent pool, creating a free-rider problem.

    “First, the Cognitive Commons is collectively dependent.Organizations in any domain rely on hiring experts from a shared talent pool, creating a free-rider problem: they benefit from available expertise while each has incentive to avoid bearing the cost of developing it, free-riding on others' developmental investments.”
    Quote in the paperSubstantial claim

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

    Interpretationreasoning

    If AI achieves fully autonomous reliability, the collective action problem shifts from expertise depletion to workforce displacement, a qualitatively different challenge.

    “if AI achieves fully autonomous reliability, the collective action problem shifts from expertise depletion to workforce displacement, a qualitatively different challenge.”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: AI systems will continue to require human oversight for reliable performance in consequential domains.

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

    Interpretationnone given

    The Cognitive Commons framework applies when developmental pathways are difficult to replicate outside organizational work contexts.

    “Second, those developmental pathways must be difficult to replicate outside organizational work contexts.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Automation creates limited opportunities for developing deep expertise.

    “Ardichvili (2022) warned that automation creates limited opportunities for developing deep expertise, but the field has not developed frameworks to understand this as a collective resource problem requiring governance at the professional level.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Ardichvili, 2022 source

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

    Interpretationreasoning

    The framework generates testable predictions that professions should show greater commons disruption where AI exposure indices are higher, where early-career developmental pathways have historically been concentrated in organizational rather than educational contexts, and where the three commons characteristics are most pronounced.

    “This occupational specificity generates testable predictions: professions should show greater commons disruption where AI exposure indices are higher, where early-career developmental pathways have historically been concentrated in organizational rather than educational contexts, and where the three commons characteristics are most pronounced.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    AI-era work demands two forms of expertise: Internalized Mastery, which is deep domain knowledge from sustained practice, and Distributed Mastery, which is fluency in orchestrating human-AI systems.

    “Understanding how this tragedy unfolds requires recognizing that AI-era work demands two forms of expertise: Internalized Mastery, representing deep domain knowledge concentrated within individual minds through sustained deliberate practice, and Distributed Mastery, representing fluency in orchestrating intelligence distributed across human-AI systems.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The Cognitive Commons is degradable through overexploitation of its regeneration mechanism: organizations can extract value from AI deployment faster than the commons can regenerate.

    “Third, the commons is degradable through overexploitation of the regeneration mechanism.Organizations can extract value from the commons (by deploying AI that depends on expert validation while eliminating the entry-level positions that train validators) potentially faster than the commons can regenerate.”
    Quote in the paperSubstantial claim

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

    Interpretationnone given

    The Cognitive Commons framework applies when AI performs tasks that historically served as primary developmental contexts for the expertise in question.

    “First, AI must perform tasks that have historically served as primary developmental contexts for the expertise in question, specifically the early-career cognitive labor and production tasks through which novices build domain knowledge, tacit understanding, and mental models.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Profession-level depletion of expertise is a structural prediction from the assisted-performance/independent-capability dissociation and commons theory, not an observed outcome.

    “The profession-level depletion this paper describes is a structural prediction from that dissociation and commons theory, not an observed outcome, and it is offered as a falsifiable account of where current incentives lead rather than a claim that the tragedy has already arrived.”
    Quote in the paperSubstantial claim

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

    Interpretationnone given

    The Cognitive Commons is the collective pool of deep human expertise within a profession, functioning as a shared resource whose regeneration can no longer be taken for granted.

    “This paper introduces the Cognitive Commons as the collective pool of deep human expertise within a profession that functions as a shared resource whose regeneration can no longer be taken for granted.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    Where formal education currently substitutes effectively for workplace experience in building relevant expertise, commons vulnerability is moderated.

    “Where formal education currently substitutes effectively for workplace experience in building the relevant expertise, commons vulnerability is moderated.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Market forces reward efficiency optimization and penalize organizations that bear disproportionate costs for expertise development whose benefits accrue profession-wide.

    “Market forces reward efficiency optimization and penalize organizations that bear disproportionate costs for expertise development whose benefits accrue profession-wide.”
    Quote in the paperSubstantial claim

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

    Interpretationreasoning

    Software engineering exhibits different commons vulnerability than plumbing, and within software engineering, backend infrastructure development may differ from front-end development depending on which tasks AI substitutes most readily.

    “Software engineering exhibits different commons vulnerability than plumbing, and within software engineering, backend infrastructure development may differ from front-end development depending on which tasks AI substitutes most readily.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    In Wiles et al.'s experiment, AI assistance during a skill-building task improved performance during the access period but did not transfer to later unassisted performance.

    “Wiles et al.'s (2024) experimental evidence is instructive: when AI assistance was provided during a skill-building task, participants performed significantly better during the access period, but this performance advantage did not transfer to subsequent unassisted performance.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Wiles et al., 2024 source

  • Classified as probably not a substantial claim.

    Interpretationnone given

    The framework does not address professions facing wholesale disintermediation, where AI eliminates the coordination or translation function the profession exists to perform rather than depleting its expertise pipeline.

    “It does not address professions facing wholesale disintermediation, where AI eliminates the coordination or translation function the profession exists to perform rather than depleting its expertise pipeline; these represent a distinct phenomenon requiring separate analysis.”
    Quote in the paperSubstantial claim

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

    Interpretationreasoning

    Apprenticeship traditions, guild structures, and community-based knowledge transmission represent alternative arrangements that may exhibit different commons dynamics or different vulnerability to AI substitution.

    “Apprenticeship traditions, guild structures, and community-based knowledge transmission represent alternative arrangements that may exhibit different commons dynamics or different vulnerability to AI substitution.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    The Cognitive Commons framework applies when the profession exhibits collective dependence on a shared expertise pool, non-excludability of access, and vulnerability to regeneration failure as organizations eliminate entry-level positions.

    “Third, the profession must exhibit the three commons characteristics: collective dependence on a shared expertise pool, non-excludability of access to that pool, and vulnerability to regeneration failure as organizations eliminate entry-level positions.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Automation simultaneously increases the need for human skill and reduces the opportunity to develop it.

    “Bainbridge (1983) demonstrated that automation simultaneously increases the need for human skill and reduces the opportunity to develop it, an insight extended to system-level analysis by Hollnagel and Woods (2005).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Bainbridge, 1983 source
    • cites Hollnagel and Woods, 2005 source

  • May combine two claims.

    Interpretationreasoning

    AI eliminates this productive activity, creating a second-order market failure: not merely under-investment in training, but elimination of the productive activity through which expertise development naturally occurred as a byproduct of employment.

    “AI eliminates this productive activity, creating a second-order market failure: not merely under-investment in training, but elimination of the productive activity through which expertise development naturally occurred as a byproduct of employment.”
    Quote in the paperNot a substantial claim

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

    Premisenone given

    Classical human capital theory assumes that despite under-investment, some training occurs because firms need trained workers to perform tasks, and the work itself provides developmental opportunities.

    “Classical human capital theory assumes that despite under-investment, some training occurs because firms need trained workers to perform tasks, and the work itself provides developmental opportunities.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    Work modularization-the degree to which professional work can be unbundled into discrete AI-compatible tasks versus requiring integrated domain judgment across the entire workflow-affects commons vulnerability.

    “Fifth, work modularization: the degree to which professional work can be unbundled into discrete AI-compatible tasks versus requiring integrated domain judgment across the entire workflow.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Medicine and engineering exhibit higher regulatory intensity and safety criticality that create institutional counterpressures, suggesting slower but not absent degradation dynamics.

    “Medicine and engineering exhibit higher regulatory intensity and safety criticality that create institutional counterpressures, suggesting slower but not absent degradation dynamics.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The six-node causal chain makes visible the reinforcing feedback loops and time delays that keep individual actors from recognizing their collective impact on the regeneration mechanism.

    “The six-node causal chain (Figure 1) formalizes these relationships, making visible the reinforcing feedback loops and time delays that prevent individual actors from recognizing their collective impact on the regeneration mechanism.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Hiring needs of organizations and the regeneration of professional expertise were linked by the operational need for junior labor, and AI breaks that link.

    “The accidental alignment argument describes structural coupling: organizational hiring needs and professional expertise regeneration were coupled through the operational necessity of junior labor, and AI decouples them.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Brynjolfsson et al. analyzed payroll data covering over 25 million U.S. workers, with an analysis sample of 3.5 to 5 million workers per month, examining employment trends by age and AI exposure after generative AI adoption.

    “Brynjolfsson et al. (2025) analyzed payroll data from a provider covering over 25 million U.S. workers, drawing on an analysis sample of 3.5 to 5 million workers per month, examining employment trends by age and AI exposure following widespread generative AI adoption.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Brynjolfsson et al., 2025

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

  • Classified as probably not a substantial claim.

    Premisecitation

    Becker distinguished firm-specific human capital, valuable only to the current employer, from general human capital, transferable across firms.

    “Becker's (1964) seminal work distinguished between firm-specific human capital (valuable only to the current employer) and general human capital (transferable across firms).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Becker, 1964

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

  • May combine two claims.

    Interpretationreasoning

    Natural resource commons regenerate once harvesting stops, but professional expertise does not, because its developmental requirements are specific and cannot be substituted.

    “Natural resource commons regenerate when harvesting stops.Professional expertise does not, because the developmental requirements are specific and non-substitutable.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Natural resource commons regenerate once harvesting stops.

  • May combine two claims.

    Interpretationreasoning

    Organizations need expertise held back for validation, crisis response, and novel situations that go beyond what AI can do, yet they lack enough economic incentive to keep that reserve because costs fall on them individually while benefits spread across the ecosystem.

    “Organizations need expertise held in reserve for validation, crisis management, and novel situations that exceed AI capabilities.”
    • Also stated in: § Why Rational Organizations Deplete the Commons: The Overgrazing Mechanism · plain
      “But they lack sufficient economic incentive to maintain that reserve when costs fall on them individually while benefits distribute across the ecosystem.”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Interpretationreasoning

    Five factors determine an occupation's commons vulnerability: task substitutability, regulatory intensity, safety criticality, professional association strength, and work modularization.

    “The framework identifies five factors that determine an occupation's commons vulnerability.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationcitation

    The evidence presented captures early-stage dynamics in a specific subset of highly AI-exposed sectors.

    “The evidence presented here captures early-stage dynamics in a specific subset of highly AI-exposed sectors.”
    Quote in the paperSubstantial claim

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

    Interpretationreasoning

    Because all organizations reason in the same way, the regeneration mechanism erodes while appropriation from the shrinking pool continues.

    “As all organizations reason identically, the regeneration mechanism erodes while appropriation from the shrinking pool continues unabated.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: All organizations reason in the same way about eliminating entry-level positions.

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

    Premisenone given

    Widespread generative AI adoption is extremely recent, with most organizational implementation beginning in late 2022.

    “Widespread generative AI adoption represents an extremely recent development, with most organizational implementation beginning in late 2022.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    The majority of professions show no measurable disruption to traditional hiring patterns.

    “The majority of professions show no measurable disruption to traditional hiring patterns.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    Occupations with strong professional associations capable of enforcing developmental standards have institutional infrastructure for commons stewardship that others lack.

    “occupations with strong professional associations capable of enforcing developmental standards have institutional infrastructure for commons stewardship that others lack.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    Professions where expertise failures produce immediately visible catastrophic consequences create accountability structures that resist degradation of substantive validation capability.

    “professions where expertise failures produce immediately visible catastrophic consequences create accountability structures that resist degradation of substantive validation capability.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    In the most AI-exposed occupations, measurable disruption to the regeneration mechanism may be emerging.

    “In the most AI-exposed occupations, measurable disruption to the regeneration mechanism may be emerging.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Software engineering, financial analysis, and legal research exhibit high task substitutability, relatively low regulatory intensity compared to medicine, and work modularization that facilitates AI substitution, placing them in higher vulnerability categories.

    “Software engineering, financial analysis, and legal research exhibit high task substitutability, relatively low regulatory intensity compared to medicine, and work modularization that facilitates AI substitution, placing them in higher vulnerability categories.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Systematic reviews of career and technical education show such programs improve initial employment but have no measurable impact on long-term earnings, suggesting workplace developmental contexts drive expertise formation.

    “Systematic reviews of career and technical education confirm the pattern: such programs improve initial employment but show no measurable impact on long-term earnings (Lindsay et al., 2024), suggesting that workplace developmental contexts, not educational preparation, drive expertise formation.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lindsay et al., 2024

      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), S4 (based).

    Interpretationnone given

    Surface validation capability can develop through training and experience with AI systems themselves.

    “This capability can develop through training and experience with AI systems themselves.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    Internalized Mastery requires cognitive struggle during formation and cannot be acquired through observation or AI-assisted performance alone.

    “Requires cognitive struggle during formation; cannot be acquired through observation or AI-assisted performance alone.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    The Validation Tether operates through a critical asymmetry: AI systems can perform tasks that appear correct, but humans need deep domain knowledge to recognize when superficially coherent outputs contain substantive errors.

    “This mechanism operates through a critical asymmetry: while AI systems can perform tasks that appear correct, humans require deep domain knowledge to recognize when superficially coherent outputs contain substantive errors.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    Surface validation is insufficient for professional work, where errors often hide beneath plausible surfaces.

    “Surface validation represents important quality control but proves insufficient for professional work where errors often hide beneath plausible surfaces.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    Surface validation allows substantively incorrect outputs to pass undetected.

    “Permits substantively incorrect outputs to pass undetected.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Knowledge that matters economically is mostly dispersed and bound to specific circumstances of time and place, and it resists being captured in any centralized representation.

    “Hayek (1945) argued that economically consequential knowledge is largely dispersed and tied to particular circumstances of time and place, resisting capture in any centralized representation.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Hayek (1945)

      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). May combine two claims.

    Interpretationnone given

    Stock depletion becomes numerically measurable through cohort analysis over 10 to 20 year periods, while functionality degradation may show up in validation error rates on shorter timescales.

    “These failure modes are distinct in timing: stock depletion becomes numerically measurable through cohort analysis over 10 to 20 year periods, while functionality degradation may manifest in validation error rates on shorter timescales.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    The degradation of the Validation Tether is invisible to performance metrics that track output volume rather than output validity.

    “Its degradation is invisible to performance metrics that track output volume rather than output validity.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    AI assistance lets junior workers who remain employed achieve productivity without cognitive struggle, so augmentation without internalization weakens expertise formation (Mechanism 2).

    “Second, AI assistance enables junior workers who remain employed to achieve productivity without cognitive struggle, Mechanism 2, through which augmentation without internalization attenuates expertise formation.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Ekuma's systematic review identified a significant lack of longitudinal studies on AI's lasting impact on HRD processes and outcomes.

    “Ekuma's (2024) systematic review identified critical gaps, noting the significant lack of longitudinal studies investigating AI's lasting impact on HRD processes and outcomes, and calling for research tracking these effects over time (p.218).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ekuma (2024) source

  • May combine two claims.

    Interpretationcitation

    Workers who become fluent in AI orchestration without deep domain knowledge gain productivity on routine tasks while losing the substantive validation capability that prevents errors on non-routine tasks.

    “Workers who develop fluency in AI orchestration without acquiring deep domain knowledge gain productivity on routine tasks while losing the substantive validation capability that prevents errors on non-routine tasks.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    Distributed Mastery is productive output achievable without deep domain internalization, and it does not supply the independent domain competence needed for substantive validation.

    “Productive output achievable without deep domain internalization.”
    • Also stated in: § An Integrative Model of Commons Depletion · plain
      “Does not provide the independent domain competence required for substantive validation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Among elite consultants, productivity gains appeared on tasks within AI capability frontiers, but performance declined on tasks outside those boundaries, exactly when substantive validation became critical.

    “Productivity gains materialized on tasks within AI capability frontiers, but performance declined on tasks outside those boundaries, precisely when substantive validation became critical.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Dell'Acqua et al. (2026) source

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

    Interpretationreasoning

    The Validation Tether is the fundamental dependence of effective AI orchestration on independent domain understanding.

    “The Validation Tether represents the fundamental dependence of effective AI orchestration on independent domain understanding.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The participants knew something was wrong but could not articulate what or why, leaving them unable to override the algorithmic output.

    “The participants knew something was wrong but could not articulate what or why, leaving them unable to override the algorithmic output.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Interpretationnone given

    The Validation Tether manifests at two distinct validation levels, surface and substantive.

    “The tether manifests at two distinct validation levels.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    The Cognitive Commons is non-rivalrous in access but collectively degradable, unlike individual human capital.

    “Unlike individual human capital (Becker, 1964), non-rivalrous in access but collectively degradable.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Becker, 1964

      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).

    Interpretationnone given

    Internalized Mastery differs from Distributed Mastery in providing the independent domain foundation required for substantive validation.

    “Distinguishes from Distributed Mastery in providing the independent domain foundation required for substantive validation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Both pathways reduce Internalized Mastery formation across successive cohorts while appearing to organizational accounting as efficiency gains.

    “Third, both pathways reduce Internalized Mastery formation across successive cohorts while appearing, to organizational accounting, as efficiency gains.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    Substantive validation requires Internalized Mastery and degrades alongside it across successive practitioner cohorts.

    “Requires Internalized Mastery; distinguishes competent human oversight from its performance.Degrades alongside Internalized Mastery across successive practitioner cohorts.”
    Quote in the paperNot a substantial claim

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

    Interpretationnone given

    Surface validation is accessible without Internalized Mastery and compatible with Distributed Mastery alone.

    “Accessible without Internalized Mastery; compatible with Distributed Mastery alone.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    The Cognitive Commons becomes a recognizable phenomenon only when commons theory, HRD scholarship, systems thinking, and distributed cognition are combined.

    “The Cognitive Commons emerges as a recognizable phenomenon only when these literatures combine.”
    Quote in the paperNot a substantial claim

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

    Interpretationcitation

    The professional-association level is the scale at which Ostrom's governance research is most directly relevant.

    “The professional-association level represents the scale at which Ostrom's (1990) governance research is most directly relevant.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ostrom, 1990 source

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

    Interpretationreasoning

    These AI-related competencies are necessary but insufficient if HRD is to steward expertise ecosystems.

    “These are necessary competencies, but insufficient if HRD is to steward expertise ecosystems.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationnone given

    Direct position elimination operates when AI systems autonomously perform entry-level tasks, allowing organizations to eliminate those positions entirely.

    “The first mechanism, direct position elimination, operates when AI systems autonomously perform entry-level tasks, allowing organizations to eliminate those positions entirely.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductreasoning

    The observed patterns are concentrated in the most AI-exposed occupations and do not characterize professional employment broadly.

    “The patterns concentrate in the most AI-exposed occupations and do not characterize professional employment broadly.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    The data cover less than three years of widespread generative AI adoption, reflecting early-stage rather than mature dynamics.

    “The data span less than three years of widespread generative AI adoption, representing early-stage rather than mature dynamics.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    Critical HRD scholars have long recognized the tensions between performance and learning, individual and collective, and efficiency and resilience.

    “Critical HRD scholars have long recognized these tensions (Bierema & Callahan, 2014;Fenwick, 2004), but the Cognitive Commons makes them unavoidable for all HRD practitioners, not just those adopting critical stances.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Bierema & Callahan, 2014 source
    • cites Fenwick, 2004 source

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

    Interpretationreasoning

    The Cognitive Commons reveals that the performance and learning paradigms sometimes pull in genuinely contradictory directions.

    “The Cognitive Commons reveals that these paradigms sometimes pull in genuinely contradictory directions, and professionals need preparation for that tension.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    By eliminating entry-level positions, organizations pursue efficiency while socializing the costs of diminishing knowledge.

    “By eliminating entry-level positions, organizations prioritize efficiency while socializing the costs of diminishing knowledge.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    Sustaining a commons first requires recognizing that one exists, understanding the collective action challenges that threaten it, and coordinating responses that overcome free-rider incentives.

    “Sustaining a commons first requires recognizing that one exists, understanding the collective action challenges threatening it, and coordinating responses that overcome free-rider incentives.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Ostrom (1990) source

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

    Premisecitation

    Ostrom's empirical work identified eight design principles characteristic of long-enduring self-organized commons institutions.

    “Her empirical work identified eight design principles characteristic of long-enduring self-organized commons institutions: clearly defined boundaries, proportional equivalence between benefits and costs, collective choice arrangements, monitoring, graduated sanctions (typically operating as reputational or peer consequences rather than state enforcement in self-organized arrangements), conflict”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ostrom, 1990 source

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

    Interpretationreasoning

    HRD can play a central role in addressing the cognitive commons challenge.

    “HRD can play a central role in addressing this challenge.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    As practitioners abandon disciplined validation habits, the conditions for intergenerational transfer of a substantive validation culture erode along with formal developmental pathways.

    “As practitioners abandon disciplined validation habits, the conditions for intergenerational transfer of substantive validation culture erode alongside the formal developmental pathways.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Junior workers using AI reach productivity levels that historically required years of experience, letting organizations accomplish the same work with fewer entry-level employees.

    “Junior workers using AI achieve productivity levels that historically required years of experience, enabling organizations to accomplish the same work with fewer entry-level employees.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    As retirees leave without adequate successors, the pool of experts shrinks, reducing the capacity for independent validation and crisis management.

    “As retirees leave without adequate successors, the pool of experts shrinks, reducing the ability for independent validation and crisis management.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Commons depletion increases systemic fragility, with failures arising from validators who lack the Internalized Mastery needed for substantive oversight and from a loss of meaningful agency in automated work.

    “This leads to increased systemic fragility, where failures occur due to validators lacking the Internalized Mastery required for substantive oversight and a loss of meaningful agency in automated work.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Surface validation works on what AI can articulate, whereas substantive validation requires tacit, contextual knowing that articulation cannot capture.

    “Surface validation operates on what AI can articulate; substantive validation requires the tacit, contextual knowing that articulation cannot capture.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Polanyi (1966)

      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).

    Premisenone given

    Nations depend on domestic expertise pools for functions that cannot be outsourced during a crisis, including critical infrastructure, healthcare, and cybersecurity.

    “Nations depend on domestic expertise pools for functions that cannot be outsourced during crisis, including critical infrastructure, healthcare, and cybersecurity.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Ostrom showed that commons can be sustained under appropriate governance, often through arrangements that arise from local experimentation rather than central design.

    “Ostrom (1990) demonstrated that with appropriate governance, commons can be sustained, often through arrangements that emerge from local experimentation rather than central design.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Ostrom (1990) source

  • May combine two claims.

    Interpretationreasoning

    The Tragedy of the Cognitive Commons is not inevitable in principle, but under current incentive structures and theoretical frameworks it is structurally predictable.

    “The Tragedy of the Cognitive Commons is not metaphysically inevitable, but it is structurally predictable given current incentive structures and theoretical frameworks.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    Daley's formal model identifies structural levers that could in principle moderate commons depletion: subsidies to organizations maintaining developmental infrastructure, metrics that reward quality over volume, and arrangements that sustain human oversight in AI-intensive work.

    “Daley's formal model identifies structural levers that could in principle moderate commons depletion: subsidies to organizations maintaining developmental infrastructure, metrics that reward quality over volume in evaluating professional output, and arrangements that sustain human oversight in AI-intensive work.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Daley (2025) source

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

    Premisecitation

    The field has lacked a framework for treating AI's constraint on deep expertise development as a collective action problem requiring profession-level governance rather than organizational-level training solutions.

    “Ardichvili (2022) warned that automation limits opportunities for developing deep expertise, but the field has lacked a framework for understanding this as a collective action problem requiring profession-level governance rather than organizational-level training solutions.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ardichvili (2022) source

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

    Premisecitation

    Ostrom's central finding was that these arrangements emerged through local experimentation and adaptation rather than external imposition.

    “Her central finding was that these arrangements emerged through local experimentation and adaptation rather than through external imposition (Ostrom, 1990(Ostrom, , 2010)).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ostrom, 1990 source

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

    Interpretationnone given

    Distributed Mastery is a legitimate and valuable form of professional expertise that requires its own developmental infrastructure and standards.

    “Governance arrangements must also recognize that Distributed Mastery is a legitimate and valuable form of professional expertise requiring its own developmental infrastructure and standards.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    The evidence does not demonstrate widespread validation failures or profession-wide expertise collapse.

    “The evidence does not demonstrate widespread validation failures or profession-wide expertise collapse.”
    Quote in the paperSubstantial claim
100 of 100

  • The authors state this as an assumption.

    Professional expertise develops through entry-level employment in hierarchical organizations and is governed through formal credentialing institutions.

    “The framework also assumes a particular cultural model of professional formation, in which expertise develops through entry-level employment in hierarchical organizations and is governed through formal credentialing institutions.”

  • The authors state this as an assumption.

    AI systems will continue to require human oversight for reliable performance in consequential domains.

    “The framework also assumes that AI systems continue to require human oversight for reliable performance in consequential domains;”

  • The paper relies on this without support.

    All organizations reason in the same way about eliminating entry-level positions.

    “As all organizations reason identically, the regeneration mechanism erodes while appropriation from the shrinking pool continues unabated.”

  • The paper relies on this without support.

    Natural resource commons regenerate once harvesting stops.

    “Natural resource commons regenerate when harvesting stops.”

  • The paper relies on this without support.

    AI systems have identifiable competence envelopes that practitioners can assess.

    “the deep domain understanding that enables practitioners to assess whether a given problem falls within or outside the AI system's competence envelope.”

  • The paper relies on this without support.

    The Cognitive Commons is currently healthy and functional across most professional domains.

    “The Cognitive Commons remains healthy and functional across most professional domains today.”
6 of 6

Cites

Flagged first, then by citations

Research works printed but not linked: 8

  1. Becker G. S. (1964). Human capital: A theoretical and empirical analysis, with special reference to education. National Bureau of Economic Research.
  2. Benzing M., Garcia C., Bermejo Cano M. (2025). 2025 agentic teaming & trust research report. Microsoft. https://aka.ms/agentresearch2025_ch1download
  3. Hayek F. A. (1945). The use of knowledge in society. The American Economic Review, 35(4), 519–530.
  4. Lindsay J., Hughes K., Dougherty S. M., Reese K., Joshi M. (2024). What we know about the impact of career and technical education: A systematic review of the research. American Institutes for Research. https://cteresearchnetwork.org/resources/2024-systematic-review
  5. McCall M. W., Lombardo M. M., Morrison A. M. (1988). The lessons of experience: How successful executives develop on the job. Lexington Books.
  6. Pigou A. C. (1920). The economics of welfare. Macmillan.
  7. Polanyi M. (1966). The tacit dimension. Doubleday.
  8. Van Manen M. (1990). Researching lived experience: Human science for an action sensitive pedagogy. State University of New York Press.

Cited by

Sorted by citations

Authors
Lovett, N.
Affiliations
North Atlantic Treaty Organization (BE)
Corresponding author
Lovett, Nolan
ORCID
Lovett, Nolan
Venue
Human Resource Development Review
Publisher
SAGE Publications
Published
2026-07-26
ISSN
  • 1534-4843 (ISSN-L)
  • 1552-6712
DOI
10.1177/15344843261470602
arXiv
2607.29380
OpenAlex
W7171305923
Semantic Scholar
afac14c80dcb3373a6feb6ed7b292837ca552d01
Cited by
7
References
54
Citations by year
2026: 7
FWCI
412.81
Type
journal article
Language
en
Open access
Licence
Locations
  • Human Resource Development Review (publishedVersion)
  • arXiv (Cornell University) (acceptedVersion, public-domain)
  • ODU Digital Commons (Old Dominion University) (acceptedVersion)
  • arXiv (Cornell University) (public-domain)
Topics
  • Social Sciences › Human Factors and Ergonomics › Innovative Education and Learning Practices
  • Social Sciences › Psychology › Applied Psychology › Human Resource Development and Performance Evaluation
  • Social Sciences › Business, Management and Accounting › Organizational Behavior and Human Resource Management › AI and HR Technologies
Keywords
artificial intelligence, professional expertise, commons theory, distributed cognition, human-AI collaboration, human oversight of AI, occupational vulnerability, professional associations, workforce policy, human resource development

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review