A Rising Tide of AI Adoption—But Not for Everyone
Microsoft’s latest Global AI Diffusion report shows that generative AI is rapidly moving into the mainstream, yet not on equal terms. Global usage climbed from 16.3 percent to 17.8 percent of the working-age population, indicating that AI tools are shifting from experimental toys to everyday work and learning companions. Twenty-six economies now report more than 30 percent of people aged 15 to 64 using AI. Beneath this headline growth, however, lies a stark AI adoption gap. Some high-income, digitally mature economies now see close to half—or more—of their working-age populations using generative AI, while others lag far behind. This uneven spread is redefining who can access AI-enabled productivity gains, professional development, and new types of employment. As AI becomes embedded in everything from search and messaging to coding and content creation, the divide between those plugged into these tools and those left out is becoming a structural feature of the global digital divide.
The Global North–South AI Adoption Gap
The clearest sign of a new global digital divide is the widening disparity between AI usage in the Global North and Global South. In the latest quarter, 27.5 percent of people in the Global North used generative AI, up from 24.7 percent just months earlier. In the Global South, usage rose from 14.1 percent to 15.4 percent—growth, but at barely half the pace. The gap widened from 10.6 to 12.1 percentage points, underscoring how developing markets AI adoption is not catching up but falling further behind. Microsoft links this divergence not to a lack of tools, but to basic infrastructure and human capital: reliable electricity, affordable and fast internet connectivity, and baseline digital skills. Without these foundations, even free or low-cost AI products remain out of reach. This compounds existing inequalities in education and work, as advanced economies layer AI capabilities on top of already-strong digital ecosystems.
AI Skills Shortage: When Talent Becomes a Luxury
As generative AI spreads, access to AI skills is emerging as a form of economic privilege. In many developing markets, AI skills shortages are more acute because education systems, corporate training pipelines, and local tech communities have had less time and fewer resources to adapt. Even where individuals are eager to learn, inconsistent connectivity and limited exposure to advanced tools create barriers that wealthier economies can more easily overcome. This imbalance risks turning AI fluency into a luxury good, concentrated among workers and firms in high-adoption economies. Meanwhile, the Global South faces the prospect of competing in global markets where productivity, software development, and knowledge work increasingly assume AI-augmented capabilities as a baseline. Without targeted investment in training, infrastructure, and local language support, developing economies could find their workforce locked out of the highest-value roles in AI-enhanced industries.
Innovation, Coding, and the New Productivity Frontier
AI’s impact is especially visible in software development, where coding tools from providers such as GitHub Copilot are reshaping workflows. Microsoft reports that global Git pushes increased 78 percent year over year, while new Git repositories rose 45 percent compared with the same quarter a year earlier. Merged GitHub pull requests linked to AI coding agents have grown more than 28 times since mid-2025, a sign that AI-assisted development is rapidly becoming a standard practice. These trends amplify the AI adoption gap. Economies with strong infrastructure and AI skills can harness these tools to accelerate innovation, build new products, and upskill developers. Others risk being confined to low-value segments of the digital economy. As AI improves in more languages and across tasks—from professional exams to everyday learning—the question is whether developing markets can secure the connectivity, training, and policy support needed to join this new productivity frontier rather than be left as passive consumers of others’ AI innovations.