Insights from interviews with policymakers and experts around the world
With the second Global AI Safety Summit in Seoul and the AI for Good Global Summit in Geneva taking place within a week, the international focus on AI is unmistakable. Over the past few months, Access Partnership has been reaching out to key policymakers and experts from countries around the world as part of our Global AI Insights project. AI is poised to be a global game changer, presenting both vast opportunities and significant risks that captivate worldwide interest. In the words of Alexandr Wang, founder of Scale AI, “Which countries have access to AI technologies and how they use them are going to define how the world plays out over the course of the next few decades.”
But even as leading policymakers gather to discuss AI policy, it is important to remember that AI is both global and local. A lot of media attention has focused on the big tech players. However, one of AI’s key characteristics is its capacity to revolutionize individual companies by optimizing operations, scaling up customer engagement, driving innovation tailored to specific market niches, and giving small businesses the automation and analysis tools they need to punch above their weight and compete with larger enterprises. AI will also accelerate and amplify how we communicate while changing our approach to tackling universal social challenges, such as healthcare, climate change, and logistics. Even as AI propels global progress and connectivity, it must be adapted to local contexts.
To ensure that our Global AI Insights project captures a diverse range of priorities, Access Partnership made a deliberate effort to reach out to the Global South and other smaller countries that often get overlooked in global public discourse. By encompassing a diverse array of viewpoints, we discerned both commonalities and differences across various regions, thereby enriching our insight into the global AI landscape and supporting a more inclusive dialogue. On the surface, almost all countries are broadly aligned. Each agrees on the need for balance and a fundamental level of interoperability. Overall, their visions encompass the following three priorities: AI Safety and Risk Management, Fostering AI Innovation, and Equitable Participation in AI. But while no country would forgo any of the three altogether, a deeper conversation reveals patterns in the relative importance and emphasis placed on each one.
AI Safety and Risk Management: Countries in the EU (which recently passed its AI Act) and others, including Australia and Mexico, had major concerns about AI safety and tended to focus on the need for risk management. The flipside of AI’s potential benefits is concerns about unintended consequences. Explicit examples of AI misuse include deepfakes and other misinformation, as well as highly sophisticated cyberattacks and fraud. There are also more subtle harms, such as the bolstering of inherent biases and the perpetuation of discrimination. A key subtopic that emerged repeatedly in our conversations was the fear that a global lowest common denominator for AI safety would not provide sufficient protection for their citizens.
Fostering AI Innovation: From our interviews, we found that countries in East Asia, the Middle East and Northern Africa, focused more on the need to foster AI innovation. The rise of AI presents significant economic opportunities to enhance competitiveness and create new goods, services, and business models. However, the countries most concerned about this priority recognize the potential for AI to help resolve significant societal challenges, such as an aging population, climate change, and healthcare. These issues often also serve as local political flashpoints. Many policymakers in these regions also tend to look past the AI media hype, worrying that the industry is still young and that regulation might be premature.
Equitable Participation in AI: In our conversations, we found the foremost concern of most countries in South America, Africa, and others, like Cambodia, to be the need to ensure more equitable participation in AI so that it can be used to address socioeconomic disparities, rather than amplify them. In the global discourse surrounding AI, it’s easy to overlook a crucial sentiment: that AI’s emergence is not occurring in isolation but against the backdrop of our existing global inequalities. AI development is resource-intensive, requiring significant computational power and vast amounts of data and connectivity. Most of all, however, it demands specialized expertise and talent. Two key subtopics that emerged include the significant challenges in attracting and developing AI talent and the proliferation and potentially uneven distribution of job losses resulting from AI.
All three priorities are critical, and there is no point in trying to argue that one should be prioritized above the others. At the local level, AI’s emergence will be moulded around local characteristics. Finding the right balance between these priorities will differ from country to country. However, while some domestic and unilateral actions can bolster AI safety or innovation, there are significant limitations to what a single country can achieve in promoting more equitable participation in AI. Global collaboration is vital for all three themes, with a multilateral approach necessary to enact meaningful change.
Access Partnership will be releasing a follow-up report that dives deeper into key topics arising from our conversations with leaders and experts from around the world. If you are a policymaker or AI business leader and would like to talk to us, you can reach us at [email protected].
