Gorde:
| Egile nagusia: | |
|---|---|
| Formatua: | Recurso digital |
| Hizkuntza: | |
| Argitaratua: |
Zenodo
2026
|
| Sarrera elektronikoa: | https://doi.org/10.5281/zenodo.20236284 |
| Etiketak: |
Etiketa erantsi
Etiketarik gabe, Izan zaitez lehena erregistro honi etiketa jartzen!
|
Aurkibidea:
- <p><span>Corporate boards are increasingly expected to oversee artificial intelligence strategy, risk, and ethics - yet most directors lack the technical literacy to evaluate AI systems and the governance frameworks to exercise meaningful oversight. This paper examines the widening gap between board-level AI governance expectations and actual board capabilities, drawing on published survey data from independent directors across publicly listed companies and analysis of corporate governance disclosures. The paper identifies three failure modes in board AI governance: performative oversight (boards approve AI strategies they cannot evaluate), delegation without accountability (AI governance is pushed to management committees without board-level follow-through), and risk blindness (boards focus on AI opportunity while systematically underweighting AI-specific risks including bias, hallucination, adversarial vulnerability, and regulatory exposure). Drawing on agency theory, stewardship theory, and the corporate governance literature on board competency, the paper proposes the Board AI Governance Competency Framework (BAGCF) - a structured model that maps specific AI oversight responsibilities to required knowledge domains and recommends structural mechanisms to close the governance gap. These mechanisms include AI-literate board advisors, technology subcommittees with external expert members, mandatory AI risk reporting cadences, and board education programs calibrated to director learning needs rather than technology vendor marketing. The paper concludes with implications for board directors exercising AI oversight, corporate secretaries structuring governance processes, institutional investors evaluating board AI competency, and director training organizations designing AI governance curricula.</span></p>