The United Nations digital technology agency has launched a new initiative aimed at improving trust in artificial intelligence agents. The program, announced this week, focuses on establishing standards and benchmarks to ensure AI systems operate transparently and ethically. The initiative seeks to address growing concerns over AI reliability and accountability in various sectors.
The initiative involves collaboration among international stakeholders, including governments, technology developers, and civil society organizations. It will develop guidelines and frameworks to assess AI agents' behavior and decision-making processes. The agency emphasized the importance of transparency and user trust as central pillars of the project, aiming to foster responsible AI adoption globally.
This move comes amid increasing scrutiny of AI technologies worldwide, with regulators and users demanding clearer accountability mechanisms. The initiative aligns with broader efforts to regulate AI and mitigate risks such as bias, misinformation, and privacy violations. Comparable efforts include the European Union’s AI Act and other national policies targeting trustworthy AI deployment.
The UN digital agency plans to release its first set of trust benchmarks for AI agents by the end of the year, providing measurable criteria for developers and regulators. These benchmarks will serve as a reference point for evaluating AI systems’ compliance with ethical and operational standards, marking a concrete step toward standardized AI governance.