WTO: AI related commodity trade measures' quantity restrictions' increase year by year

The hot topic of artificial intelligence (AI) is how to ensure that everyone can benefit from it while reshaping the global economic and trade landscape.

Recently, the "2025 World Trade Report Seminar and Chinese Version Release Conference" was held during the 8th China International Import Expo. Attendees expressed that AI is developing rapidly and has enormous potential, but its inclusiveness should be taken seriously to ensure equal benefits for all countries.

Johanna Hill, Deputy Director General of the World Trade Organization (WTO), stated in a video message that AI has enormous potential to drive global trade and gross domestic product (GDP) growth. According to the WTO Secretariat's forecast, by 2040, artificial intelligence is expected to drive nearly 40% growth in global trade in services and 12% to 13% growth in the global economy.

This report, titled "Synergizing Trade and AI for the Benefit of All Parties," focuses on the interactive relationship between AI and international trade, analyzing their impact on inclusive growth and the direction of international cooperation. This report is the annual report of the WTO.

According to the latest outlook from the WTO in October, AI related commodity trade is leading strong growth in the first half of 2025. Among them, the trade growth rate of AI related goods in the first quarter of this year was 16.5%, which increased to 21.7% in the second quarter; The trade growth rates of non AI related goods during the same period were only 3.7% and 4.2%, respectively.

AI and trade mutually empower each other

According to WTO data, the total global trade volume of AI related goods was $2.9 trillion in 2022, slightly decreasing to $2.3 trillion in 2023. Since 2012, the import of AI related goods has significantly increased, mainly driven by intermediate products such as computer components and testing equipment, such as evaluation kits, graphics cards, memory modules, and cooling systems. Compared to the promotion of "computers, semiconductors, and other equipment" and "raw materials/processing chemicals", "intermediate goods" account for the largest share and grew the fastest between 2017 and 2022, followed by a decline in 2023. Regarding the reasons for the decline, the WTO believes that it may be due to global trade restrictions, changes in regulation and export capacity, or previous strategic inventory accumulation.

The export of AI related products is mainly concentrated in the Asian region represented by China and South Korea, with significant efforts in intermediate inputs and equipment. Although the European Union and the United States are also major exporters of AI related intermediate inputs and equipment, their export growth rates are relatively moderate. Emerging manufacturing centers such as Malaysia, Mexico, Vietnam, and Thailand have also increased their exports of AI related intermediate inputs and equipment in recent years.

In the first half of this year, driven by structural investments in digital infrastructure such as semiconductors and cloud servers, the trade volume of AI related goods increased by 20% year-on-year. Although it only accounts for less than 10% of the global total commodity trade, it accounts for nearly half of the global year-on-year growth. Compared to AI products, the trade performance of non AI goods increased by less than 4% year-on-year.

There are also differences in the contribution of AI related commodity trade among different regions. In terms of imports, North America has performed outstandingly, with semiconductor imports increasing by 36%. However, the majority of the growth in AI related trade comes from Asia, accounting for nearly two-thirds of the total AI related trade growth in the first half of 2025. Asia has shown strong export performance in artificial intelligence related products, which is consistent with the surge in global investment in this field.

In the view of the WTO, AI has become an important catalyst for trade driven growth. For example, AI can reduce trade costs by optimizing supply chains, automating customs clearance, lowering language barriers, and promoting industry wide productivity improvement.

The report emphasizes how AI can help businesses reduce costs in three major areas: logistics, regulatory compliance, and communication. A joint survey by the WTO and the International Chamber of Commerce (ICC) in 2025 shows that 70% of businesses expect the use of AI to reduce trade costs, with small and medium-sized enterprises (with 249 or fewer employees) being more optimistic in this regard compared to larger enterprises (with over 250 employees). Nearly 90% of companies using AI have achieved tangible benefits in trade activities, and 56% of companies believe that AI has improved their trade risk management capabilities.

In the three major fields mentioned above, 44% of the surveyed small and medium-sized enterprises believe that AI can help companies reduce costs by at least 1/4 in the logistics field, and the same is true in the regulatory compliance field. In terms of communication, 37% of surveyed small and medium-sized enterprises expect AI to reduce costs in this field by more than half. For large enterprises, only in the field of communication, 46% of the surveyed large enterprises expect AI to reduce costs in this area by about 1/4, and only 11% expect to reduce costs by more than half.

According to WTO simulations, by 2040, AI may drive a 34% to 37% increase in global trade. Specifically, digitally deliverable services have the highest growth rate of 42%, mainly due to reduced trade costs, high tradeability of AI services, and expansion of high productivity industries. Meanwhile, AI may drive global GDP growth of 12% to 13%.

Tariffs and the Challenge of Digital Divide

In early April this year, US President Trump implemented so-called "equivalent" tariffs globally, disrupting the global trade order. In this context, the WTO has lowered the expected growth rate of global commodity trade volume from 2.8% in 2024 to 2.4% in 2025 and 0.5% in 2026. The forecast for global GDP growth remains at 2.7% in 2025 and 2.6% in 2026. Although the commodity trade forecast for 2025 has been raised compared to the forecasts for April and August, the WTO's downward adjustment of expectations for 2026 indicates that the impact of current US tariff policies will further shift to 2026.

WTO economists emphasize that the main downside risk facing current forecasts is the spread of trade restrictions and policy uncertainty to more economies and industries. Of course, on the positive side, the sustained growth of trade in artificial intelligence related goods and services may boost global trade in the medium term.

At the same time, the report also points out that in the field of trade policy, although tariffs on AI related goods have been relatively low in recent years, non-tariff measures such as export restrictions and technological barriers are increasing year by year. According to WTO data, there will be nearly 500 "quantity restrictions" (QR) on AI related goods worldwide in 2024, and their proportion among all "quantity restriction" measures is also on the rise, approaching 18%. As the main form of non-tariff barriers, quantity restrictions refer to administrative measures taken by a country (region) government to regulate the import and export quantity of a certain commodity within a certain period of time (usually one year). The problem of fragmented data governance is prominent, with over 80% of the restrictive measures in 452 cross-border data policies coming from high-income and middle to high-income economies, increasing compliance costs for enterprises.

In addition to the aforementioned impacts, the current integration of AI and trade also faces significant imbalances, especially in digital infrastructure. The report shows that low-income economies are far behind high-income economies in terms of Internet penetration, broadband speed and other indicators. By 2024, there will still be 2.6 billion people worldwide who are unable to connect to the internet, mainly concentrated in developing regions. The indicator of "AI application" shows a "centralized" feature, that is, more than 60% of large enterprises in high-income economies use AI, while the proportion in low-income economies is less than one-third; At the industry level, the application rate of AI intensive fields such as finance and IT services reaches 52% to 61%, while the penetration rate of AI in manufacturing is less than 25%.

Regarding bridging the digital divide, the WTO believes that international cooperation is a key breakthrough point. The WTO has previously issued standards such as the Information Technology Agreement and the General Agreement on Trade in Services, aimed at providing a foundation for AI related trade. In the future, the WTO plans to expand the coverage of the Information Technology Agreement and reduce tariff constraints on key AI raw materials; Promote data rule coordination and reduce localization requirements; Strengthen collaboration with international organizations, such as collaborating with the International Organization for Standardization (ISO) to develop AI standards, and collaborating with the International Labour Organization (ILO) to address labor market shocks.