Canada could lead on AI governance

Artificial intelligence (AI) constitutes a general-purpose technology, and perhaps one with transformative potential not seen since the industrial revolution. While the precise speed at which AI technology will advance is unknown, its enormous power to spur both positive and negative socio-economic change is already obvious. Now is the time to understand our collective interests on AI, and to find ways to build governance that is in the human and global interest. There is an opportunity for Canada to lead internationally at different levels and from different orientations. This includes government, research communities, civil society, and the private sector. Canada needs to take leadership now as it builds towards hosting the G7 in 2025. Advances in user-friendly generative AI, which can produce text, images, audio, and synthetic data, have unleashed a frenzy of new initiatives and captured mainstream attention. The best example: ChatGPT reached about 100 million users in two months—an unprecedented feat that took TikTok around nine months and Instagram two-and-a-half years to achieve. The popularity of ChatGPT and its checkered performance (it can generate convincing, but completely incorrect, material at times) have opened the floodgates of debate on the benefits and risks associated with AI. It has raised longstanding ethical and legal concerns related to privacy and harmful biases. It has also created news ones. No policy areas—from democratic integrity, labour, education, market competition, to art—will remain untouched. Beyond domestic policy, AI will become increasingly central in geopolitics and international co-operation, changing the worlds of finance, climate change, trade, development, and military operations and security. Paul Samson is president of the Centre for International Governance Innovation. Photograph courtesy of CIGI AI has also become central to national security and industrial policy planners at the heart of government. China’s AI plan is to become the undisputed global leader in technology and research by 2030, while the U.S. National Artificial Intelligence Initiative seeks to bolster its own continued leadership. Many elements of both strategies foster conditions for direct competition. Moreover, the CHIPS for America Act aims to secure U.S. AI research and development and supply chains, and China appears to be initiating security reviews of foreign tech companies for cyber risks to ensure its supply chains. The raw material for AI is data, and yet the current patchwork of regulations scattered across nations do not provide the necessary foundation for AI governance. The EU is considering an ambitious AI Act with rules for data and AI, but the arrival of generative AI may delay moving forward. Both the U.S. Algorithmic Accountability Act and the White House's Blueprint for an AI Bill of Rights have uncertain timelines. Other countries, including Japan, Canada, Brazil, and South Korea, are also reviewing national AI legislation. Over the past year, China implemented nationally binding regulations in targeted areas. Alibaba’s new chatbot tool has prompted Chinese regulators to require companies to submit generative AI products (e.g., online chat tools) for assessment before public release. India, the G20 president in 2023, is leading the way on digital public infrastructure, but it will be very challenging to make progress on governance. What options do policymakers have on global governance for AI? The G7 will discuss AI again at the Hiroshima Summit this year, and the Japan presidency is hopeful to agree to an international framework for AI governance. Under a real stretch scenario, the G7 may find agreement on a framework for international AI governance and build support with the G20, and beyond. But it is difficult, even for the G7, to reach agreement in high-level discussions among ministers and leaders on such a technical topic. A helpful bottom-up approach could be to establish a G7 leader-empowered digital stability board, or another informal, technical-driven group to convene multi-stakeholder expertise to prioritize a set of standards for AI and ensure alignment with the latest technology. No new organizations or reformed mandates are needed for this step. Existing bodies could be leveraged, including standards development organizations and state-led bodies, like the Forum for Information and Democracy, the EU-U.S. Trade and Technology Council, and the Global Partnership on AI, which was launched during the previous French and Canadian G7 presidencies. Regardless of which approach is taken, though, developing countries must be central to AI governance models, or else digitization and AI will further increase inequalities between nations, as well as within them. The timing is perfect now for Canada to step up on a critical domestic and global issue where there is an opportunity to lead at home, and abroad. Paul Samson is president of the Centre for International Governance Innovation. The Hill Times