The G20’s 2025 discussions mark a decisive moment in global AI development. For the first time, the G20 Digital Economy Ministers gathered in Cape Town, South Africa, reaffirming a shared responsibility to ensure that technological progress strengthens human rights, expands opportunity and supports the Sustainable Development Goals. Their commitment reflects a global understanding that AI is reshaping economies and societies, but must do so responsibly.

AI emerged as one of the strongest priorities, with leaders recognizing its power to boost economic growth, enhance public services, and enable solutions to global challenges. The G20 underscored that safe, secure, and trustworthy AI requires strong governance, robust datasets, interdisciplinary research, and AI ecosystems that empower every country, especially developing nations.

The push for ethical AI builds on earlier frameworks such as the G20 AI Principles (2019) and the 2024 Maceió Declaration, expanding them into practical cooperation. Member countries agreed on risk-based, human-centric governance aligned with human rights, privacy protections, and intellectual property laws. They emphasized the need for transparency, accountability, representative datasets, and guarding against algorithmic bias.

Digital divides remain a major global obstacle. Ministers highlighted that disparities in computing resources, talent development, and local language datasets disproportionately impact the Global South. To address this, the G20 called for stronger capacity building, policy sharing, open-source AI adoption, and infrastructure investments including data pools and interoperable data-sharing mechanisms.

The rapid rise of generative AI intensified discussions on safety. Members raised concerns about deepfakes, misinformation, non-consensual content, online manipulation, and identity fraud. Children, women, elderly populations, and people with disabilities are especially vulnerable. The G20 explored policy responses, including watermarking of AI-generated content, strengthening digital literacy, and improving technical detection systems.

Global cooperation emerged as a recurring theme. The G20 welcomed UN resolutions on safe and secure AI and capacity building, emphasizing that developing countries must be meaningfully included in global governance. Multistakeholder collaboration across governments, companies, researchers, and civil society is seen as essential for advancing equitable AI, strengthening AI research, and ensuring assessment tools that evaluate AI performance throughout its lifecycle.

By the close of discussions, members emphasized that AI’s future must reflect inclusivity, fairness, sustainability, and shared prosperity. Ethical AI is no longer a theoretical ideal but a global priority shaping the next phase of digital transformation.

Companies like Gates AI are helping translate these commitments into practical progress. Through its Responsible AI certification, governance toolkits, and technical assurance frameworks, Gates AI supports governments and institutions working to build safe, transparent, and inclusive AI systems. Its focus on accountability, evaluation, and bias mitigation reflects the same priorities emphasized by the G20, reinforcing a shared mission to ensure that AI development benefits all societies and advances global progress.