The verification bottleneck is developed as a distinct socio-technical mechanism and queueing epistemics is introduced, a framework for analysing knowledge reliability when verification-demanding outputs arrive faster than bounded review capacity can process them, to establish a general design principle: AI productivity should be governed by verification capacity, not generation capacity alone.
Kwan-Hong Tan· Open Access Journal of Multi...· 0 citations
Enterprise adoption of agentic artificial intelligence (AI) is moving from passive text generation toward autonomous planning, tool use and cross-system workflow execution. This transition creates a control gap: conventional model governance evaluates outputs or development processes, while agentic systems create risk through sequential actions, delegated authority and changing operating context. This paper develops a runtime assurance architecture (RAA) for enterprise agentic AI and formalizes a quantitative Autonomy-Risk Exposure (ARE) score for deciding when an agent may execute, must be sandboxed, requires human approval or must be blocked. A design-science method was used to synthesize requirements from AI risk-management standards, generative AI security guidance and agentic AI threat literature. The model was then evaluated through a reproducible scenario simulation of 2,000 enterprise agent episodes across knowledge assistance, data retrieval, internal workflow and external transaction tasks. Results show that the full RAA configuration reduced mean ARE from 35.7 to 24.4 points (31.5% reduction), decreased invalid or policy-conflicting actions from 9.8% to 4.6%, eliminated unsupervised pass-through of high-risk invalid actions in the simulated environment, and improved mean audit evidence coverage from 0.61 to 0.89. The control benefit was achieved with a mean latency overhead of 95 ms and human approval for 12.6% of episodes. The paper contributes a practical reference architecture, a risk-scoring equation, a policy decision algorithm and implementation guidance for organizations deploying agentic AI in regulated or high-consequence workflows.
K. Tan· World Journal of Advanced Re...· 1 citation
Generative artificial intelligence can increase the speed and apparent quality of knowledge work, yet conventional evaluations rarely determine whether users remain capable of understanding, verifying, contesting, remembering and taking responsibility for AI-mediated outputs. This paper develops the Cognitive Sovereignty Threshold (CST), an interdisciplinary humanities and sciences framework for distinguishing sovereign augmentation from cognitively fragile efficiency. The study uses integrative conceptual synthesis and design-science modelling to connect extended cognition, cognitive offloading, automation reliance, metacognition, epistemic agency, narrative responsibility and institutional governance. Cognitive sovereignty is operationalised through five dimensions: epistemic authorship, verification capacity, metacognitive calibration, contestability and retention. A geometric aggregation model is combined with a delegation-oversight penalty to produce a Cognitive Sovereignty Index (CSI), while a Sovereignty-Adjusted Value measure links task performance to retained human agency. Seven transparent analytic scenarios illustrate how similar productivity levels can conceal sharply different sovereignty profiles. AI used as an adversarial critic or verified drafting partner produces the strongest joint performance and sovereignty outcomes, whereas opaque, mandatory or answer-first use produces fragile efficiency even when immediate task performance appears high. The paper introduces the principle of germane cognitive friction: human-AI systems should deliberately preserve the effort required for source inspection, counterargument, reason-giving, delayed recall and meaningful override. The framework contributes a testable construct, a formal threshold model and a practical audit architecture for education, organisations and public institutions. It concludes that responsible AI adoption should optimise not only output quality and risk controls, but also the continued human capacity to know, judge, explain and act without compulsory dependence on the system.
Kwan-Hong Tan· International Journal of Res...· 1 citation
Agentic artificial intelligence (AI) is transforming social and organisational life by moving AI from an assistive tool into a semi-autonomous actor that can plan, recommend, communicate, coordinate workflows and trigger decisions. Existing debates on trustworthy AI often emphasise technical reliability, legal compliance or ethical principles, but social adoption depends on a subtler question: how do people learn when to trust, distrust, verify or refuse AI outputs in everyday work? This conceptual research develops the Social Trust Calibration Framework (STCF) for responsible human-AI adoption in organisations and public institutions. Using an integrative conceptual synthesis of trust theory, automation research, organisational sensemaking, algorithmic management, human-centred AI and contemporary AI governance guidance, the study identifies the mechanisms through which AI trust becomes either calibrated, excessive, deficient or displaced. The resulting framework argues that trust in agentic AI must be calibrated across four mutually dependent layers: capability trust, process trust, institutional trust and identity trust. The paper contributes a formal definition of calibrated AI trust, a social calibration loop, a diagnostic matrix of trust states, a maturity model and seven research propositions for future empirical testing. The analysis shows that appropriate reliance cannot be achieved by accuracy alone. It requires visible evidence, role-specific verification routines, accountable decision rights, psychological safety, transparent escalation paths and protection of human agency. The paper concludes that responsible AI adoption should be treated as a social trust calibration problem rather than a simple technology acceptance problem. This reframing offers practical guidance for leaders, educators and policymakers seeking to scale AI while preserving human judgement, legitimacy and social confidence.
K. Tan· International journal of res...· 0 citations
Enterprise agentic artificial intelligence (AI) increasingly converts model outputs into consequential actions involving payments, records, customer communications, infrastructure, and regulated decisions. Existing safeguards commonly emphasize refusal, confidence thresholds, expected loss, or human approval, but they insufficiently distinguish a recoverable task failure from an irreversible or externally propagated harm. This paper develops Reversibility-Aware Staged Delegation (RASD), a multidisciplinary decision framework integrating AI governance, resilience engineering, transaction processing, real-options reasoning, and human-centered automation. RASD introduces an Action Recoverability Index, Non-Recoverable Exposure, and an option-value decision rule that allocates each proposed action among direct execution, staged commit, human review, and block/defer modes. The staged mode separates preparation, validation, commitment, and compensation so that an agent can make progress while preserving the organization’s ability to inspect, reverse, or contain side effects. A formal dominance condition shows when staging creates greater expected value than direct execution. The framework is evaluated in a Monte Carlo design comprising 120,000 synthetic enterprise tasks across 240 episodes, including a controlled distribution shift. RASD achieved a mean net value of 7.408 normalized units per task, compared with 5.735 for a confidence-threshold policy and 5.282 for an expected-loss gate. Its severe-incident rate was 0.390%, versus 5.937% and 4.166%, respectively, while preserving positive value after distribution shift. RASD had a higher raw task-failure rate than the expected-loss gate, demonstrating that failure frequency alone is an inadequate safety metric when recovery and consequence containment differ. The findings support a shift from binary autonomy decisions toward recoverability-preserving execution architectures and provide operational guidance for auditability, human escalation, and risk-adjusted enterprise value creation.
K. Tan· Open Access Journal of Multi...· 0 citations
An original theoretical model is developed to explain why the same agentic AI capability can generate measurable value in one organization but produce negligible or negative returns in another, and that sustainable AI value does not increase monotonically with either automation intensity or governance intensity.
K. Tan· International Journal of Sci...· 1 citation
This qualitative study conducts a comparative document analysis of ten influential governance instruments issued by UNESCO, the OECD, the European Union, the Council of Europe, the United States National Institute of Standards and Technology, the United Kingdom, the Group of Seven, and Singapore.
Kwan-Hong Tan· Open Access Journal of Multi...· 0 citations
The paper argues that algorithmic governance should not be assessed only by whether systems are accurate, explainable or compliant, but also by whether affected persons retain interpretive agency, contestatory power, relational recognition and meaningful participation in institutional life.
K. Tan· International Journal of Law...· 0 citations
The ESG-digital nexus is developed as a strategic management framework explaining how digital capabilities can strengthen ESG performance and how ESG objectives can discipline digital investment toward long-term value creation.
K. Tan· International Journal of Sci...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
