These guidelines from AlgorithmWatch outline ethical principles for using generative AI, including transparency, accountability, and safeguards against harm, with a focus on practical policy and governance measures for organizations.
Why it matters for ENACT-AI
The resource develops a critical and responsible approach to generative AI by emphasising that AI outputs should not be accepted without scrutiny. It addresses misinformation, bias, framing, missing perspectives, data protection, privacy, environmental and social risks, human oversight, and transparency. It is particularly valuable for developing responsible AI practices and understanding that AI use requires human judgement and accountability.
Learning outcomes
Learners will be able to identify key risks and benefits associated with generative AI, critically evaluate AI-generated content, recognise bias, framing, missing perspectives, and inaccuracies in AI outputs, distinguish between different levels of data sensitivity, understand the importance of privacy and security when using AI tools, apply quality assurance principles, explain the importance of transparency in AI-assisted work, and make more informed decisions about when and how generative AI should be used.
Suggested activity
Ask participants to develop a responsible AI decision framework for their organisation, school, or youth group. Present several realistic AI use cases, such as generating research ideas, summarising confidential documents, translating content, creating public communication, or analysing personal data. Participants classify each use case according to proportionality, security, quality assurance, and transparency. They then decide whether AI should be used, under what conditions, and what human checks are required. Finish with a group discussion about situations where not using AI may be the more responsible choice.