What are the best practices for engaging with stakeholders on AI and human rights issues?

1 Identify your stakeholders

Your stakeholders are anyone who can affect or be affected by your AI project, such as your customers, employees, partners, suppliers, regulators, investors, civil society, and the public. You need to identify who they are, what their interests and concerns are, and how they can influence or be influenced by your AI project. You also need to consider the potential impacts of your AI project on different groups of people, especially those who are vulnerable or marginalized, and ensure that their voices are heard and respected.

2 Communicate your purpose and values

You need to communicate clearly and openly with your stakeholders about the purpose and values of your AI project, and how they align with the human rights standards and norms. You need to explain what your AI project does, how it works, what data it uses, what benefits it provides, what risks it entails, and how you mitigate them. You need to use simple and accessible language, avoid jargon and technical terms, and provide examples and scenarios that illustrate your points. You also need to listen and respond to your stakeholders' feedback, questions, and suggestions, and show that you value their input and perspectives.

3 Involve your stakeholders in the design and development process

You need to involve your stakeholders in the design and development process of your AI project, and seek their input and participation at every stage. You need to consult them on the objectives, methods, criteria, and indicators of your AI project, and ensure that they reflect the human rights principles and values. You need to co-create and co-test your AI project with your stakeholders, and incorporate their feedback and recommendations into your iterations and improvements. You also need to evaluate and monitor the impacts of your AI project on your stakeholders, and report and disclose the results and outcomes.

4 Establish a governance and accountability mechanism

You need to establish a governance and accountability mechanism for your AI project, and ensure that it is transparent, independent, and effective. You need to define the roles and responsibilities of the different actors involved in your AI project, such as the developers, users, providers, and overseers, and ensure that they adhere to the human rights standards and norms. You need to create and implement policies and procedures for the ethical use, oversight, and audit of your AI project, and ensure that they are accessible and understandable for your stakeholders. You also need to provide and enable mechanisms for redress and remedy for any adverse impacts or harms caused by your AI project, and ensure that they are fair and responsive.

5 Learn from others and share your experiences

You need to learn from others and share your experiences on engaging with stakeholders on AI and human rights issues, and seek to collaborate and cooperate with other actors in the field. You need to follow and adopt the best practices and guidelines from various international frameworks and initiatives, such as the UN Guiding Principles on Business and Human Rights, the OECD Principles on AI, the EU Guidelines on Trustworthy AI, and the Global Network Initiative. You also need to join and contribute to the relevant networks and platforms, such as the Partnership on AI, the Responsible AI Forum, and the AI Global Governance Commission, and exchange knowledge and insights with other innovators, experts, and practitioners.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?