Public sector workers using GenAI tools made available by their organisation disclose when content is AI-generated, particularly in communications and interactions with the public, make that clear to recipients and communicate the outputs' limitations such as factual inaccuracies or missing context. They avoid entering sensitive, proprietary or personal data and comply with data protection law; always validate and verify outputs before use in decisions or communications; treat the tool as complementing not replacing judgement, critically assessing all outputs and applying human review and approval in critical or sensitive areas; review outputs for bias and harmful stereotypes, with post-processing filters where appropriate; and check the use case against the NCSC's GenAI cyber security guidance for public sector bodies.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.