Diagnosis of the challenges and opportunities

The successful integration of trustworthy Artificial Intelligence (AI) in everyday life requires taking into account a complex interplay of challenges, opportunities, and governance considerations, each influencing the trajectory and impact of AI technologies on society. For this to happen, effective governance, regulation, and knowledge of its mechanisms are crucial to foster AI innovation while ensuring an AI culture that it conforms to ethical norms and societal values. Although AI innovation can thrive in a well-regulated environment that encourages ethical development and deployment for advancing trustworthy AI, this requires governments, industry leaders, academic institutions, and civil society to be aware about how AI achieves outcomes and why, while considering differences in each nation’s regulations, technological capabilities, and regional digital divides.

Our vision

The integration of trustworthy AI into our societal frameworks necessitates a fundamental shift from rational to relational thinking about personhood, data, justice, and everything in between, placing ethics as something that transcends mere technical solutions. To achieve this, AI technologies need to be developed and deployed not just for efficiency and operational gains but also for promoting transparency, justice, safety, responsibility, privacy, autonomy, self-determination, sustainability and cultural diversity. This means embracing a relational approach that prioritizes the human and societal impact of AI, thus recognizing the interconnectedness of individuals, communities, and technologies while aligning with three core values: human dignity, democracy, and the rule of law.

Core values

• Human dignity: AI systems must respect and enhance human dignity, ensuring that individuals are treated with respect and their fundamental rights are upheld. Therefore, we must advance to develop AI that empowers people and improves their quality of life (i.e. human-centered AI).
• Democracy: AI must support democratic processes and institutions, promoting transparency and a participatory approach involving many relevant stakeholders that further advance responsible AI technology for the general public interest, thus adapted to context-specific social value requirements.
• Rule of law: AI development and deployment must be governed by and adhere to the rule of law, ensuring that all activities are conducted in accordance with the law to ensure justice and accountability in all AI-related endeavors.

Key commitments

• Ethical development: We must commit to the ethical development of AI technologies, integrating ethical considerations into every stage of the AI lifecycle. This includes addressing biases, ensuring fairness, and prioritizing the well-being of all affected individuals.
• Transparent governance: We must commit to clear and transparent governance frameworks for AI, ensuring that policies and regulations with respect to AI are openly communicated and subject to public scrutiny. This transparency is essential for building trust and legitimacy.
• nclusive innovation: We must commit to create opportunities and resources for diverse participation in AI innovation. This includes supporting education and training initiatives, promoting diversity in AI research and development, and ensuring that all voices are heard.
• Accountability and responsibility: We must commit to implement robust accountability mechanisms to monitor and evaluate AI systems, ensuring that they operate as intended and do not cause harm. Responsibility for AI outcomes needs to be clearly defined and enforced.
• Sustainable development: We must commit to sustainability in AI development and use in order to minimise the environmental impact of AI while promoting their use in addressing global challenges such as climate change and social inequality.

Recommendations

By fostering an environment where AI is developed and governed in line with the abovementioned core values and key commitments, we can move closer to realizing the vision of "better" AI that supports social justice and contributes to the broader goals of the 2030 Agenda for Sustainable Development. For this purpose, we propose the following 5 recommendations

1. To develop and continuously update ethical frameworks for AI that are sensitive to the diverse societal contexts in which AI operates. These frameworks should be dynamic to accommodate evolving societal norms, technological advancements, and regulatory changes. They should also be informed by broad stakeholder engagement, including input from marginalized and vulnerable communities, to ensure that the frameworks are inclusive and representative of diverse societal needs and values.
2. To implement participatory governance models that actively involve a wide range of stakeholders in the decision-making processes related to AI development and deployment. This should include policymakers, technologists, civil society organizations, and, crucially, the end-users of AI technologies, especially those from underrepresented groups. Such engagement will help ensure that AI systems are developed and governed in a manner that is transparent, accountable, and aligned with the public interest, thereby enhancing trust and acceptance among all societal sectors.
3. To encourage and support interdisciplinary collaboration between technologists, social scientists, ethicists, and other relevant stakeholders to explore the socio-technical complexities of AI. This approach should aim to integrate qualitative and quantitative research methods to gain a holistic understanding of the impacts of AI on society. Interdisciplinary collaboration can foster the development of AI technologies that are not only technically advanced but also ethically sound, culturally sensitive, and socially beneficial, ultimately contributing to the achievement of sustainable development goals.
4. To implement comprehensive ethical and impact assessments throughout the lifecycle of AI systems, from design to deployment and beyond. These assessments should evaluate the potential social, economic, and cultural impacts of AI technologies, with a particular focus on identifying and mitigating risks related to bias, discrimination, and inequality. By making these assessments an integral part of the AI development process, organizations can proactively address ethical concerns and ensure that AI systems contribute positively to society and do not inadvertently harm vulnerable populations.
5. To develop educational programs and public awareness campaignsto enhance AI literacy among all segments of society, including policymakers, developers, and the general public. These initiatives should cover the ethical dimensions of AI, the potential benefits and risks of AI technologies, and the rights of individuals in the context of AI decision-making. By fostering a well-informed public discourse on AI, these efforts can empower individuals to be able to harness the power of AI in a well-considered and responsible manner, to engage in meaningful discussions about AI ethics, to advocate for their rights, and to participate actively in shaping the development and governance of AI technologies in a way that aligns with democratic values and social justice principles.

Call to Action

For this vision of trustworthy AI to become a reality, the abovementioned core values need to be embraced, the key commitments realised and the key recommendations respected by various stakeholders such as governments, industry leaders, academic institutions, and civil society. Moreover, this vision extends to utilizing AI to address region-specific challenges, strengthen local industries, and improve the quality of life for all residents. This will ensure that the development, deployment, and governance of AI technologies benefit society while mitigating potential harms and addressing ethical, social, and environmental concerns. Together, we can create a future where AI technologies are not only transformative but also aligned with the values that define us as humanity. Thus, let us work together to ensure that AI serves the common good, enhances human dignity, supports democratic principles, and upholds the rule of law.

Signatories

We, the undersigned, commit to advancing trustworthy AI as outlined above