As policy officers, AI governance officers, procurement officers, legal advisers and data protection officers in public administration, we sit at critical points of decision-making on the deployment of artificial intelligence (AI) in our administrations. This includes the procurement of AI systems and services, the use of AI in public services and administrative decisions affecting residents, and digital strategy decisions that determine how far public services come to depend on external AI providers.
Existing AI literacy and governance training does not equip us to assess the safety of advanced AI systems, and we currently lack the funding to upskill the people who hold similar positions.
We therefore sign this statement to show our support for Katalina Hernández in creating an AI safety training and upskilling programme specifically for public administration officers, covering:
- how advanced AI models are built and evaluated, and how they fail;
- assessing the claims and documentation of AI vendors and developers;
- procuring and contracting for AI, including where liability sits;
- keeping critical public services from becoming dependent on a single AI provider;
- human oversight and auditability of AI-assisted decisions.
We sign in our professional capacity. Each of us intends to take part in the programme if it is funded.
This statement is not an exercise of any public authority or administrative power, and it does not commit public funds.