Anthropic Economic Index report: Uneven geographic and enterprise AI adoption – Anthropic

Report on AI Adoption and its Implications for Sustainable Development Goals (SDGs)

Introduction: AI’s Rapid Diffusion and the Urgency for SDG Alignment

Artificial Intelligence (AI) is being adopted at an unprecedented rate, far exceeding historical technology diffusion timelines for electricity, personal computers, and the internet. This rapid integration into economic processes presents both significant opportunities and challenges for the achievement of the United Nations Sustainable Development Goals (SDGs). While AI can drive innovation and productivity, early adoption patterns reveal a concentration that risks exacerbating existing inequalities, directly impacting global progress on the SDGs. This report analyzes current AI usage trends to assess their alignment with key development goals, particularly those concerning inequality, education, innovation, and economic growth.

Geographic Disparities in AI Adoption: A Challenge to SDG 10 (Reduced Inequalities)

Analysis of global AI usage reveals a stark geographic concentration, posing a direct threat to SDG 10 (Reduced Inequalities). The benefits of AI appear to be accumulating in high-income regions, potentially widening the economic gap between developed and developing nations and reversing recent progress in growth convergence.

• Concentration in High-Income Economies: The Anthropic AI Usage Index (AUI), which measures usage relative to a region’s working-age population, shows a strong positive correlation with national income. Technologically advanced economies like Singapore (AUI of 4.6x) and Canada (AUI of 2.9x) demonstrate significantly higher per-capita adoption than emerging economies such as India (0.27x) and Nigeria (0.2x).
• Risk of Economic Divergence: Historically, transformative technologies have led to significant divergence in global living standards. If the productivity gains from AI are primarily captured by high-adoption economies, current usage patterns suggest a risk of deepening global economic inequality.
• Digital Infrastructure and Access: Disparities in adoption are linked to factors such as digital infrastructure, economic structure, and regulatory environments, highlighting the need for targeted interventions to ensure the benefits of AI do not accrue only to already-prosperous economies, a core tenet of SDG 9 (Industry, Innovation, and Infrastructure).

Evolving AI Applications: Impacts on SDG 4 (Quality Education) and SDG 9 (Innovation)

The specific applications of AI are evolving, with notable shifts that carry implications for several SDGs. The diversification of AI use in high-adoption countries contrasts with a narrower focus in low-adoption regions, affecting progress on goals related to education and innovation.

1. Growth in Knowledge-Intensive Fields: There is a sustained increase in AI usage for educational and scientific tasks. Educational applications surged from 9.3% to 12.4% of total usage, and scientific tasks grew from 6.3% to 7.2%. This trend supports SDG 4 (Quality Education) by demonstrating AI’s potential as a tool for learning, research, and knowledge synthesis.
2. Task Specialization in Emerging Economies: Lower-adoption countries exhibit a heavy concentration on coding tasks, which account for over half of all usage in India compared to approximately one-third globally. While beneficial for specific sectors, this narrow focus may limit the broader societal benefits associated with diverse AI applications in education, science, and business operations.
3. Innovation and Infrastructure: The widespread use of AI for software development and IT-related tasks aligns with SDG 9 by fostering innovation. However, ensuring this innovation translates into inclusive and sustainable industrialization requires broadening AI applications beyond a few specialized domains.

Enterprise Deployment and the Future of Work: Implications for SDG 8 (Decent Work and Economic Growth)

The deployment of AI by enterprises, primarily through APIs, is heavily skewed towards automation, which has profound implications for SDG 8 (Decent Work and Economic Growth). While automation can drive productivity, it also presents significant challenges to labor markets and the nature of work.

• Dominance of Automation: Enterprise API usage is overwhelmingly automated, with 77% of business applications involving direct task delegation to AI, compared to approximately 50% for consumer use. This systematic deployment is a primary channel through which AI will reshape economic activity.
• Labor Market Disruption: The focus on automating tasks, particularly in coding and administrative support, may displace workers in roles susceptible to automation. This poses a risk to the goal of full and productive employment and decent work for all, necessitating proactive policies for reskilling and workforce adaptation, linking back to SDG 4.
• Productivity vs. Inclusivity: While automation can enhance productivity, a key driver of economic growth, its benefits may not be distributed equitably. The data suggests that model capabilities and economic value currently outweigh cost considerations in enterprise deployment, indicating that businesses are prioritizing high-value automation that could concentrate economic gains.

Key Barriers and Enablers for Equitable AI Diffusion

Achieving the potential of AI to advance the SDGs requires addressing key barriers to its broad and equitable adoption. Analysis of enterprise usage reveals critical factors that could either constrain or enable a more inclusive AI-driven transformation.

1. The Context Bottleneck: The effective deployment of AI for complex tasks is constrained by the availability of relevant contextual information. This implies that firms and nations lacking modernized data infrastructure may be unable to leverage AI for high-impact applications, creating a bottleneck that reinforces the digital divide and hinders progress on SDG 9.
2. Organizational Investment: Overcoming this bottleneck requires significant investment in data modernization and organizational restructuring. This highlights the need for policies that support digital literacy and infrastructure development to ensure that the capacity to deploy sophisticated AI is not limited to a few advanced firms or economies.
3. Price Insensitivity and Value Focus: The finding that enterprises prioritize model capability over cost suggests that the primary driver of adoption is the economic value of automation. For AI to contribute positively to the SDGs, its value proposition must be aligned with broader societal goals, including sustainable development and shared prosperity.

Conclusion and Policy Recommendations for SDG-Aligned AI Governance

The early patterns of AI adoption are characterized by a striking concentration in specific tasks and high-income geographies, posing a significant risk of economic divergence that could undermine the Sustainable Development Goals. The heavy emphasis on automation in enterprise deployment signals a forthcoming transformation of labor markets and economic structures. To mitigate these risks and harness AI for global benefit, policy choices are paramount. Actions taken now by policymakers, business leaders, and civil society will determine whether AI becomes a tool for widening inequality or a catalyst for achieving a more sustainable and equitable future. It is imperative to foster an ecosystem that promotes inclusive access, supports workforce transitions, and directs AI innovation towards solving humanity’s most pressing challenges as outlined in the SDGs.

Analysis of Sustainable Development Goals in the Article

1. Which SDGs are addressed or connected to the issues highlighted in the article?

1. SDG 4: Quality Education
   • The article highlights a significant increase in the use of AI for educational and scientific purposes. It notes that “Educational Instruction and Library tasks rose from 9% in V1 to 12% in V3” and “Life, Physical, and Social Science tasks increased from 6% to 7%.” This directly connects to the goal of leveraging technology to enhance learning and knowledge acquisition.
2. SDG 8: Decent Work and Economic Growth
   • The core theme of the article is the economic impact of AI, including its potential to boost productivity, reshape labor markets, and drive economic growth. It discusses how AI adoption can lead to “productivity gains” and “workforce changes.” The analysis of automation versus augmentation, where “directive conversations… jumped from 27% to 39%,” and the potential for AI to displace workers or create new forms of work, are central to this goal.
3. SDG 9: Industry, Innovation, and Infrastructure
   • The article is fundamentally about the diffusion of a new, innovative technology (AI) and the digital infrastructure that supports it. It examines the “unprecedented adoption speed” of AI and how its deployment relies on “existing digital infrastructure.” The analysis of enterprise API use and the factors driving business adoption relate directly to technological upgrading and innovation within industries.
4. SDG 10: Reduced Inequalities
   • A major concern raised in the article is the risk of AI exacerbating inequalities between and within countries. It explicitly states that “the benefits of AI may concentrate in already-rich regions—possibly increasing global economic inequality.” The data shows a stark “digital divide,” with high-income countries having much higher AI usage rates than emerging economies, as measured by the Anthropic AI Usage Index (AUI).
5. SDG 17: Partnerships for the Goals
   • The article demonstrates a commitment to this goal by promoting knowledge sharing and collaborative research. The authors state, “we have open-sourced the underlying data to support independent research on the economic effects of AI.” This action is intended to “catalyze independent research” and enable policymakers and researchers to develop evidence-based responses to AI’s impacts, fostering a multi-stakeholder partnership for sustainable development.

2. What specific targets under those SDGs can be identified based on the article’s content?

1. Target 4.4: By 2030, substantially increase the number of youth and adults who have relevant skills, including technical and vocational skills, for employment, decent jobs and entrepreneurship.
   • The article’s finding that AI usage for “Educational Instruction” and “Life, Physical, and Social Science tasks” is rising suggests that individuals are using AI to acquire knowledge and skills. The heavy use of AI for coding and other “Computer and Mathematical tasks” (36% of total usage) directly relates to the development of technical skills relevant for the modern economy.
2. Target 8.2: Achieve higher levels of economic productivity through diversification, technological upgrading and innovation, including through a focus on high-value added and labour-intensive sectors.
   • The article explores how businesses are deploying AI to automate tasks, which is a direct form of technological upgrading aimed at increasing productivity. The finding that “77% of business uses involve automation usage patterns” and that businesses prioritize AI for high-value tasks (even if they cost more) aligns with this target.
3. Target 9.c: Significantly increase access to information and communications technology and strive to provide universal and affordable access to the Internet in least developed countries.
   • While not focused on affordability, the article directly measures the disparity in access to and use of a key information and communications technology (AI). The creation of the Anthropic AI Usage Index (AUI) and the finding that emerging economies like “Indonesia at 0.36x, India at 0.27x and Nigeria at 0.2x, use Claude less” highlights the challenge of achieving equitable access to advanced technologies.
4. Target 10.2: By 2030, empower and promote the social, economic and political inclusion of all, irrespective of age, sex, disability, race, ethnicity, origin, religion or economic or other status.
   • The article’s analysis of the “uneven geography of early AI adoption” directly addresses the economic dimension of this target. The finding of a “strong positive correlation between Claude adoption and Gross Domestic Product per working-age capita” indicates that the benefits of this technology are not being equally distributed, potentially reversing “growth convergence seen in recent decades” and hindering the economic inclusion of populations in lower-income countries.
5. Target 17.6: Enhance North-South, South-South and triangular regional and international cooperation on and access to science, technology and innovation and enhance knowledge sharing on mutually agreed terms.
   • The article’s section on “Open source data to catalyze independent research” is a direct implementation of this target. By “open-sourcing the underlying data,” the authors are enhancing access to knowledge and data, enabling a global community of researchers to study the economic effects of AI and fostering international cooperation on a critical technological innovation.

3. Are there any indicators mentioned or implied in the article that can be used to measure progress towards the identified targets?

1. For Target 4.4:
   • Percentage share of AI usage for educational tasks: The article provides specific data, stating that “educational tasks surged from 9.3% to 12.4%.” This metric can be tracked over time to measure the extent to which AI is being used as a tool for learning and skill development.
2. For Target 8.2:
   • Share of automation vs. augmentation usage: The article quantifies the shift towards automation, noting that “Directive conversations… jumped from 27% to 39%” for consumers and that enterprise API usage is “77% automation dominant.” This serves as a direct indicator of technological upgrading in production processes.
3. For Target 9.c:
   • The Anthropic AI Usage Index (AUI): This is a custom indicator created in the article to “measure whether Claude.ai use is over- or underrepresented in an economy relative to its working age population.” The AUI values for different countries (e.g., Singapore at 4.6x, Nigeria at 0.2x) provide a clear, quantifiable measure of the digital divide in AI adoption.
   • Firm-level AI adoption rate: The article cites Census Bureau data showing AI adoption among US firms rising “from 3.7% in fall 2023 to 9.7% in early August 2025,” which is an indicator of technology diffusion in the industrial sector.
4. For Target 10.2:
   • Correlation between AI usage and GDP per capita: The article quantifies this relationship, stating “a 1% increase in GDP per capita being associated with a 0.7% increase in Claude usage per capita.” This statistical relationship serves as a powerful indicator of the economic inequality in technology adoption.
5. For Target 17.6:
   • Availability of open-source datasets on AI usage: The article’s action of “open-sourcing the underlying data” is a direct, albeit qualitative, indicator. The existence and comprehensiveness of such datasets can be used to measure progress in knowledge sharing on new technologies.

4. Summary Table of SDGs, Targets, and Indicators

Source: anthropic.com