Metraipolis › archive › dc_ce66358f33db

Framework for Government Policy on Agentic and Generative AI: Governance, Regulation, and Risk Management (2025) · OpenAlex W4414365039

Source: OpenAlex
Original: https://doi.org/10.36227/techrxiv.175825815.57896005/v1
License (as archived): CC-BY-4.0
Archived: 2026-10-08T02:04:21.187Z
SHA-256 of the archived text: f89d546f14d766bdc0f39ffceaa503310b7110c337dd65fea55e0fd82cc155d6
Stable URL of this page: https://metraipolis.com/documento/dc_ce66358f33db

How to cite

Framework for Government Policy on Agentic and Generative AI: Governance, Regulation, and Risk Management (2025) · OpenAlex W4414365039 — OpenAlex, https://doi.org/10.36227/techrxiv.175825815.57896005/v1 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_ce66358f33db (Metraipolis archive dc_ce66358f33db, sha256 f89d546f14d766bdc0f39ffceaa503310b7110c337dd65fea55e0fd82cc155d6)

BibTeX

@misc{dcce66358f33db,
  title = {Framework for Government Policy on Agentic and Generative AI: Governance, Regulation, and Risk Management (2025) · OpenAlex W4414365039},
  howpublished = {OpenAlex},
  url = {https://doi.org/10.36227/techrxiv.175825815.57896005/v1},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_ce66358f33db; sha256 f89d546f14d766bdc0f39ffceaa503310b7110c337dd65fea55e0fd82cc155d6},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: OpenAlex, https://doi.org/10.36227/techrxiv.175825815.57896005/v1. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Opera scientifica in accesso aperto: scheda dai metadati OpenAlex (preprint).
Titolo: Framework for Government Policy on Agentic and Generative AI: Governance, Regulation, and Risk Management
Autori: Satyadhar Joshi
Anno: 2025 · Lingua: en
DOI: https://doi.org/10.36227/techrxiv.175825815.57896005/v1
Accesso aperto: gold · testo: https://doi.org/10.36227/techrxiv.175825815.57896005/v1 · licenza del testo: cc-by
Citazioni ricevute (OpenAlex): 1
Temi (OpenAlex): Ethics and Social Impacts of AI; Digitalization, Law, and Regulation; Safety Systems Engineering in Autonomy

Abstract (testo degli autori, licenza cc-by):
The transition from Generative AI to Agentic Artificial Intelligence (AI) represents a fundamental shift in autonomous systems: from generating content to perceiving, reasoning, acting, and collaborating with minimal human input. This evolution unlocks significant economic potential, with estimates in the trillions of dollars, and introduces practical challenges across safety, security, ethics, and regulatory compliance. Evaluating current governance frameworks-including ISO/IEC 42001 and the EU AI Act-we find significant gaps in addressing continuous monitoring, human-agent collaboration, and multi-agent coordination. Based on these findings, we propose a multi-layered governance framework that integrates technical safeguards, ethical alignment, and adaptive regulatory compliance. Key pillars include Governance by Design, automated risk detection, and human oversight to mitigate both technical and societal risks. Our analysis highlights that traditional AI governance frameworks, designed for static or narrow AI models, are inadequate for dynamic, multi-agent, goal-oriented systems. We systematically characterize Agentic AI by its core capabilities, multi-step decision-making processes, and architectural components. We identify a comprehensive risk taxonomy encompassing technical failure, adversarial exploitation, emergent unsafe behaviors, and accountability gaps. Evaluating current governance frameworks-including ISO/IEC 42001 and the EU AI Act-we find significant gaps in addressing continuous monitoring, humanagent collaboration, and multi-agent coordination. Based on these findings, we summarize a multi-layered governance framework that integrates technical safeguards, ethical alignment, and adaptive regulatory compliance. Key pillars include Governance by Design, automated risk detection, and human oversight to mitigate both technical and societal risks. Our findings underscore the need for targeted policy interventions, operational best practices, and focused research on measuring and mitigating the realworld impact of Agentic AI. By bridging technical, ethical, and regulatory dimensions, this work provides actionable guidance for stakeholders aiming to deploy autonomous AI systems safely, responsibly, and effectively.

Fonte: https://openalex.org/W4414365039 — metadati OpenAlex (CC0 1.0), consultati il 2026-10-07 via api.openalex.org. Interrogazione: «AI governance regulation».