Microsoft Report: US AI Paradox—Global Innovation Leader, 24th in Adoption

In the rapidly evolving landscape of 2025, Microsoft’s latest AI Diffusion Report unveils a double-edged sword: while generative AI is permeating daily life at an unprecedented pace, it is simultaneously deepening global inequalities and reshaping international power dynamics. This is more than a collection of tech adoption metrics; it is a clear warning that without deliberate and forward-looking policy intervention, the "AI dividend" may remain concentrated within a few privileged nations and demographics.

The report, titled"Global AI Adoption in 2025: A Widening Digital Divide", released by the Microsoft AI Economy Institute in January 2026, notes that by the second half of 2025, global AI usage reached 16.3% of the world's population, up from 15.1% in the first half of the year. This implies that roughly one in every six people globally is utilizing AI for learning, working, or problem-solving. For a technology that entered the mainstream only three years ago, this rate of adoption is exceptionally rare in the history of industrial revolutions.

However, the structural imbalances behind this growth are stark. Data reveals that the growth rate of AI adoption in the Global North is nearly double that of the Global South. By the end of 2025, 24.7% of the working-age population in Northern countries had used generative AI tools, compared to just 14.1% in the South. This gap expanded from 9.8 percentage points at the start of the year to 10.6 points by its end.

This disparity is not accidental. It stems from long-standing inequities in infrastructure, digital literacy, and the availability of affordable AI models. These factors continue to limit the upward mobility of emerging markets within an AI-driven economy. Leading the pack are nations that made institutional investments in AI early on. The United Arab Emirates (UAE) remains the global leader with a 64% adoption rate among its working-age population—a result of a strategy dating back to 2017 with the appointment of the world’s first Minister of AI. Singapore follows closely at 60.9%, with Norway, Ireland, France, and Spain consistently ranking in the top tier.

The common denominator among these leaders is governance. They have not only invested in infrastructure and talent but have also built public trust in AI. According to the 2025 Edelman Trust Barometer, public trust in AI in the UAE reached 67%, whereas it stood at only 32% in the United States. This suggests that the integration of AI into society depends less on the technology itself and more on the pace of policy and the design of governance frameworks.

South Korea stands out as the most significant case study in this report. In the second half of 2025, its global ranking jumped seven spots to 18th place, with adoption rates rising from 25.9% to 30.7%. Since October 2024, South Korea has seen a growth rate of over 80%, far exceeding the global average of 35%.

This surge was driven by three converging forces: rapid institutional design (including the "AI Basic Act" and the National AI Strategy Commission) which integrated AI into education and the public sector; key breakthroughs in Korean-language model capabilities (GPT-5 achieved a perfect score of 100 on the CSAT benchmark); and the "Ghibli-style" image generation trend that swept social media in April 2025, successfully converting first-time users into long-term adopters.

This "Policy-Language-Culture" synchronization model serves as a clear reference for other regions. For instance, Taiwan’s AI adoption rate reached approximately 28.4% in late 2025, ranking 23rd globally. If Taiwan can further strengthen Traditional Chinese language models and localized application scenarios, it may accelerate AI's societal penetration.

In contrast, while the United States remains a leader in AI infrastructure and frontier research, it slipped to 24th in terms of adoption rate (28.3%). This highlights a structural paradox: technological leadership does not automatically translate into universal adoption. Without inclusive strategies, the fruits of innovation may remain locked within specific demographic silos.

Perhaps the most disruptive shift in the global landscape is the rise of China’s open-source platform, DeepSeek. By releasing models under the MIT license and providing free chatbots, DeepSeek has significantly lowered the barrier to entry, expanding rapidly across China, Russia, Iran, Cuba, Belarus, and Africa. （Related： Taiwan's Per Capita GDP Set to Exceed $40,000, Most Citizens Unaffected ｜ Latest ）

In Africa, adoption rates of these models are estimated to be 2 to 4 times higher than in other regions, an influence further amplified by partnerships with corporations like Huawei. This is no longer just a commercial competition; it is a battle for AI diffusion on a geopolitical level. Amidst intensifying US-China AI rivalry, open-source models have become tools for exporting influence, potentially shaping new technological dependencies in markets that are sanctioned or otherwise excluded from Western ecosystems.