Geopolitics and Generative AI: Will Middle Powers Reshape the Map?
PR Newswire
7 min read
BOSTON, Dec. 9, 2024 /PRNewswire/ -- As the generative AI map takes shape, the US and China are asserting their dominance. Tech companies from these GenAI superpowers have built a substantial lead in the creation and large-scale commercialization of top-performing large language models (LLMs). Another group of countries—the "GenAI middle powers"—is emerging, however, each with distinct strengths that may enable it to compete at a regional, and even global, scale as a supplier of the technology.
Boston Consulting Group logo (PRNewsfoto/The Boston Consulting Group)
A new report, "How CEOs Can Navigate the New Geopolitics of GenAI," from the BCG Henderson Institute and BCG's Center for Geopolitics, looks at the current state of play and the implications for both corporate leaders and policymakers.
"As it becomes clear that GenAI will shape industries and societies, the emergence of middle powers signifies a critical shift in the global balance of technological power," said Nikolaus Lang, global leader of the BCG Henderson Institute and a coauthor of the report. "For corporate leaders who are integrating GenAI into their business operations, relying solely on GenAI supplied by companies in the US or China could pose serious risks due to the possibility of regulations, data requirements and availability of the models all being vulnerable to shifts in government policy."
Two AI Superpowers…For Now
The two AI superpowers—the US and China—are currently the only players with robust access to and control over sizeable portions of the entire GenAI value chain. They produce the most intellectual property (IP) and they have the largest AI talent pools; they have some of the richest data ecosystems in the world and the most data center infrastructure capacity, and they lead in capital access.
The US has had a pronounced head start, building on decades of AI leadership. Nearly 70% of the world's notable AI models since 1950 have been developed by or in partnership with US-based companies or academic institutions, as have 57% of top-performing large language models (LLMs). The US is home to 60% of the top 2,000 AI scholars in the world and attracted roughly one-quarter of all AI specialists that relocated globally between 2022 and 2024; its total AI talent pool has grown to nearly half a million people, the largest in the world. US-based GenAI startups have also received unparalleled private investment: a total of $65 billion since 2019.
Some signs indicate that China is catching up to the US. Two Chinese companies (Alibaba and the GenAI startup 01.AI) contribute over one-quarter of the world's top open-source LLMs. Established tech giants Baidu and Tencent have also released high-performing models, as have a new generation of GenAI startups, the so-called AI tigers. Top Chinese LLMs have substantially reduced the gap compared to state-of-the-art alternatives in the last year. China benefits from ample data center infrastructure and a strong AI talent bench, and current limitations in China's access to cutting-edge chips for AI model training and inference are likely to delay rather than impede further progress.
Who Could Emerge as a New GenAI Player?
This US–China story has fueled a "two superpowers" narrative, but momentum is building in other parts of the world.
The European Union can position itself as a significant GenAI middle power by combining the strengths of its member states. The EU is already home to a nascent GenAI startup ecosystem and has the second-largest global AI specialist talent pool. As a sizeable market, with a combined GDP of $18 trillion, it can support the development of locally developed AI models that may be viewed as more protective of users' data in the context of its demanding regulatory environment. While its GenAI startups have attracted considerable investment, the EU faces challenges in scaling infrastructure and bridging the funding gap compared to the US and China.
Saudi Arabia and the United Arab Emirates (UAE) can rely on substantial sovereign wealth funds and focused investments to rapidly grow their AI talent bench, expand data center infrastructure, and develop high-performing LLMs. Saudi Arabia has committed $40 billion to AI development and is increasing its data center capacity, while the UAE has launched a $10 billion AI VC fund and developed high-performing Arabic language models. Both nations' efforts are supported by low energy costs and easing restrictions on access to cutting-edge chips. To sustain current advancements, Saudi Arabia and the UAE will need to continue growing their respective talent bases and expand technology export markets.
Japan and South Korea can build on their established tech ecosystems, significant R&D investments, and expertise in hardware to position themselves as GenAI middle powers. South Korea benefits from its strong position in semiconductor manufacturing and domestic companies making inroads in developing LLMs, while Japan has fostered partnerships among corporate, academic, and governmental institutions to accelerate model development. However, both nations will likely need to pursue a regional focus to consolidate sufficient market demand to justify the scale of investments needed to compete in the GenAI space.
Some countries may be able to compete by capitalizing on their robust AI research ecosystems and talent pools to build better, more cost-efficient LLMs. For example, Canada and the UK are home to a sizeable share of the world's top AI scholars, have produced many notable AI models, and have published some of the most important AI research papers. Israel, despite its smaller population, has a high concentration of AI specialists and has developed competitive models. For these countries, opportunity lies in accelerating research breakthroughs.
A Call to Action for Leaders
On the one hand, companies across industries have a vested interest in a diversified supply of GenAI. The lessons of Covid-19 are stark: a massive disruption can create severe chokepoints and destroy value. If all options originate in just two countries, it could lead to disruptions in GenAI's availability due to geopolitical shocks. Corporate leaders will need to build the "geopolitical muscle" required to sense coming shifts and adapt their operating models.
On the other hand, because AI sovereignty may soon become a critical source of national security, economic value, and soft power, governments of countries or in regions with potential to become GenAI middle powers should consider what it would take to claim the space.
"This race is far from over," said Nikolaus Lang. "Both business and political leaders will need to navigate a dynamic and high-stakes environment, shaped by regulation, policy and technology. Whether the middle powers—and which ones—will establish a claim to the competitive supply of GenAI at a global scale is one of the biggest unknowns. How this plays out has the potential to rewrite the rules of the game."
About Boston Consulting Group Boston Consulting Group partners with leaders in business and society to tackle their most important challenges and capture their greatest opportunities. BCG was the pioneer in business strategy when it was founded in 1963. Today, we work closely with clients to embrace a transformational approach aimed at benefiting all stakeholders—empowering organizations to grow, build sustainable competitive advantage, and drive positive societal impact.
Our diverse, global teams bring deep industry and functional expertise and a range of perspectives that question the status quo and spark change. BCG delivers solutions through leading-edge management consulting, technology and design, and corporate and digital ventures. We work in a uniquely collaborative model across the firm and throughout all levels of the client organization, fueled by the goal of helping our clients thrive and enabling them to make the world a better place.
