In May 2026, nearly 70 civil servants in Colombo, Sri Lanka, underwent intensive AI Literacy training, highlighting a global scramble to equip public sectors with ethical AI understanding. Organized by UNESCO with European Union support, this four-day initiative (May 18-21) aimed to build capacity for ethical decision-making and human-AI collaboration. Sri Lanka is also engaging with the Readiness Assessment Methodology (RAM) to evaluate its preparedness for ethical AI governance. Yet, despite these localized efforts, practical, comprehensive AI literacy and governance implementation remains fragmented, often overlooking critical areas. The disconnect between high-level ethical aspirations and the realities of AI deployment risks beneficial AI being overshadowed by systemic biases and unforeseen societal harms.
Defining Ethical AI and Early Sectoral Efforts
Establishing a common understanding of ethical AI is critical. 'Trustworthy AI' is generally defined by seven key requirements: human agency, technical robustness, privacy, transparency, and fairness, as outlined by digital-strategy. The framework embeds ethical considerations from design to deployment.
Sector-specific responses are emerging. In April 2026, the Ohio Board of Professional Conduct released an Ethics Guide for Lawyers and Judicial Officers. Concurrently, the Poynter Institute launched a new AI hub for journalists. While these initiatives recognize the need for ethical AI, they also reveal a fragmented landscape without a unified, adaptable ethical framework.
The Unseen Gaps in Ethical Implementation
Despite the proliferation of ethical guidelines, critical dimensions are still overlooked, especially in specialized fields. A systematic review of 17 empirical articles on AI in Education (AIED) from January 2018 to June 2023 found ethical aspects like learning analytics and algorithmic ethics often neglected in existing AIED frameworks, according to pmc. The contrast with the comprehensive 'Trustworthy AI' requirements outlined by digital-strategy is sharp. The disconnect means high-level principles fail to translate into practical, domain-specific guidelines, leaving significant ethical gaps. Many AI systems are thus deployed without adequate safeguards for vulnerable users, revealing a critical blind spot between principle and practice.
