Introduction
Over the past two centuries, humanity has experienced an unprecedented improvement in living standards. Since the Industrial Revolution began more than 200 years ago, global per capita income has increased more than tenfold. Life expectancy has nearly doubled in many countries, rising from around 40 to over 80 years in developed nations. Travel that once took months now takes hours. Communication that once depended on mail carried by horseback now happens instantaneously across continents. Today, generative artificial intelligence (GenAI) can compose symphonies, write poetry, and create artwork that rivals human creativity – capabilities that would have seemed like magic just decades ago.
These remarkable gains reflect the power of innovation and technological progress. At the core of this advancement is creative destruction – the mechanism through which successive waves of innovation replace older technologies and business models, driving long-run productivity and growth, as highlighted in the work of Nobel prize-winning economists Philippe Aghion and Howitt’s seminal work on the subject.
Yet the mere invention of new technologies tells only part of the story. Creating innovative solutions does not automatically translate into economic growth or societal benefits. For new technologies to fulfill their potential, they must be adopted and used effectively by firms and households. This so-called technology diffusion process represents a crucial bridge between invention and impactful innovation. It is neither automatic nor guaranteed.
Technology diffusion faces several challenges that can slow or prevent the spread of beneficial innovations. Users often need to acquire new skills to operate unfamiliar technologies effectively. Breakthrough technologies from the internal combustion engine to information and communication technologies require substantial investments in supporting infrastructure. For instance, consider how the motor car requires not just manufacturing plants, but networks of roads, gas stations and repair services. Businesses may need to reorganize their operations or develop new management practices. Sometimes they must create entirely new business models to harness technology’s full potential.
The arrival of breakthrough technologies typically spurs waves of complementary innovations. These organizational and business model innovations often prove as important as the original technological advance. New ways of organizing work, serving customers or structuring entire industries can generate major productivity gains that extend far beyond the technology itself. The internet, for example, enabled not just faster communication; it created entirely new forms of commerce, entertainment and social interaction that continue to reshape the global economy.
Multiple factors shape diffusion outcomes, including available skills, competitive dynamics, access to finance, and technical standards and regulations. These factors help explain why technologies do not diffuse seamlessly across economies. These uneven patterns contribute to persistent inequalities in economic development, living standards, health outcomes, and environmental quality.
Purpose and scope of this report
This year's World Intellectual Property Report seeks to shed light on the technology diffusion process. It examines what explains successful diffusion outcomes and what role intellectual property (IP) plays and how it influences the diffusion course. Intellectual property rights incentivize innovation but can also affect how quickly new technologies are diffused, creating a crucial balance for policymakers to strike.
This introductory chapter establishes the foundation for the report’s analysis by presenting the core concepts that shape technology diffusion. It reviews research examining how rapidly and extensively various technologies have spread across the globe, highlighting how diffusion patterns have changed throughout history. Building on this context, the chapter examines the key factors influencing the speed and success of diffusion.
Chapter 2 examines how technological knowledge spreads across borders – and why some economies are far better at absorbing it than others. Using patent data and citation links between inventions as its primary lens, it shows how quickly new ideas attract attention abroad, how the gap between domestic and international knowledge flows has narrowed over time, and how diffusion patterns differ across technologies and regions. The chapter also looks at how breakthrough inventions originating in one country are taken up and built upon elsewhere, revealing persistent differences in countries’ ability to identify, adapt and reuse new technologies.
Chapters 3 to 5 bring these concepts to life through three detailed case studies. They examine agricultural biotechnology, clean technologies, and digital technologies, exploring in concrete terms how diffusion processes unfold in practice. The case studies reveal the specific challenges and opportunities that arise in different technological domains. They offer insights into policy approaches that can encourage beneficial technology diffusion and illuminate the nuanced role that IP plays in shaping diffusion outcomes.
Together, these chapters aim to provide policymakers, business leaders and researchers with a comprehensive understanding of technology diffusion. This knowledge can inform decisions about innovation policy, IP systems, and strategies for harnessing technological progress to improve economic outcomes and address global challenges.
Key concepts and terminology
Technology diffusion involves multiple interconnected processes operating at different scales – from individual to industry-wide transformation. Understanding these process is important for analyzing how technologies spread and for identifying the factors that shape their outcomes. This section introduces the core concepts and the terminology that guide the rest of the report.
Technology diffusion
Technology diffusion refers to the broader spread of new technology across firms, industries, and economies as more users adopt it over time.
Diffusion of technological knowledge
The diffusion of technological knowledge refers to the spread of know-how, expertise and information that makes it possible to adopt technology. This may include scientific principles, design details, production techniques, hands-on experience, or the skills needed to operate and manage a technology in practice.
The amount of knowledge that needs to spread varies according to technology. Some technologies require very little. For instance, a farmer can apply a new fertilizer without understanding its chemistry. Others demand much more. A company that adopts advanced robotics must not only purchase the machines, but also acquire the skills required to program, maintain and integrate them into production.
These differences help explain why some technologies diffuse quickly while others spread more slowly or unevenly. The way knowledge circulates, and the extent to which users can access it, often shapes the overall diffusion process.
Technology adoption
Technology adoption describes how individuals and organizations start using a new technology and integrate it into their activities. The technology may be completely new – for example, a breakthrough from a research lab – or it may simply be new to a specific user, region or economy, even if others already use it elsewhere. Note that adoption focuses on individual cases, whereas diffusion looks at the aggregate pattern.
Adoption rarely means copying a technology in its original form. Users usually adapt it to suit the conditions. Farmers may adjust a drought-resistant crop variety to different soils or climates. Manufacturers may modify a production method when relying on different inputs, machines or business models. A digital inventory system built for large supermarket chains may need major adjustments before small corner shops can afford to use it.
These adaptation needs explain why adoption takes time and why success rates vary. They also show that adoption is not a passive act of imitation, but an active, often innovative, process.
Technology transfer
Technology transfer is a particular form of adoption. It involves the deliberate sharing of knowledge, skills, methods, or technologies between parties.
Universities transfer technology when they license research discoveries to companies. Multinational corporations transfer technology when they share production know-how or management practices with foreign subsidiaries. Governments may facilitate transfer by linking research institutions with industry or by funding joint projects.
Knowledge spillovers
Knowledge spillovers occur when ideas, expertise or skills spread from one economic actor to another without a deliberate transfer or payment. For example, a company may introduce a new product, and competitors learn from it by observing its features or by seeing how customers respond. The original company bears the cost of innovation, but other firms also benefit from the knowledge it generates.
Note that knowledge spillovers may further technology adoption outside a formal technology transfer framework. The key distinction lies in intent: technology transfer involves purposeful and often contractual sharing of knowledge, whereas spillovers happen unintentionally with others benefitting from knowledge flows that the originator did not plan nor can control.
Stylized facts
Economists have examined technology diffusion at the level of individual innovations, as well as across industries and entire economies. Although patterns and outcomes often vary depending on the technology and context, the literature reveals several stylized facts that hold more generally. These facts offer a useful lens for understanding how technologies spread and what factors determine the extent and pace of diffusion.
Stylized facts #1: Technology diffusion often follows an S-shaped path
One of the first formal studies to examine how technologies diffuse after their invention is Zvi Griliches’ pioneering 1957 work on hybrid corn in the United States.
What makes the S-shape important is not the curve itself, but what it reveals about the process of diffusion.
Once enough adopters are on board, however, a tipping point is reached, and diffusion accelerates sharply. Policies, falling prices or improvements in complementary infrastructure can all help push a technology into this rapid-growth phase.
Where do today’s key technologies sit along the S-curve? For mobile communications, the picture is one of technologies that have reached the upper, flattening part of the curve (see Figure 1.2). Coverage for 2G, 3G and now 4G has largely levelled off as these generations approach saturation. Even 5G – despite having reached half of the world’s population in under five years – will eventually slow, as the gaps that remain become harder to close. Renewable energy technologies tell a different story. Their diffusion curves still sit closer to the middle part of the S-shape, where growth is rapid and often exponential (Figure 1.3). Renewable energy generation continues to rise steeply, especially wind and solar. There is still substantial room to climb before nearing saturation, as the share of renewables in global electricity generation is projected to rise from 35 percent to 47 percent by 2030, while the share of solar and wind renewable energy sources is set to almost double to 27 percent.
While the S-curve captures an important general tendency, economists have also pointed out its limitations. Empirically, not all technologies follow a neat S-shaped trajectory. Many innovations exhibit uneven or multi-phase diffusion, with abrupt accelerations, slowdowns or successive waves, as improved versions replace earlier ones.
Stylized facts #2: Newer technologies are diffusing faster
Evidence suggests that the speed at which technologies reach widespread use has increased over time. One prominent study from the United States
In the mid-20th century United States, a product that had just begun to spread could take more than a decade to move from limited household use to near saturation. By the 1980s, this timespan had roughly halved. Innovations such as microwave ovens or personal computers reached mass adoption in half the time it had taken earlier products like washing machines or refrigerators.
The study traced this acceleration to several broad forces. Rising incomes meant that households could afford new goods sooner, while periods of economic stability with low unemployment further supported faster uptake. Demographic changes, including urbanization and evolving household structures, expanded the pool of potential adopters. The nature of the products themselves also mattered: some benefitted from infrastructure that could be quickly built or was already available, such as broadcast networks for radios and televisions. In other cases, consumers had a choice between different versions of the same product early on. Once the technology had proved useful, this variety in formats helped to lower prices and made the product easier to obtain, encouraging faster spread – as happened with videocassette recorders and personal computers.
Stylized facts #3: Across countries, technology adoption lags are shortening
Technologies are spreading around the world faster than ever before. A useful way to measure this is through adoption lags – the number of years between the first invention of a technology anywhere in the world and its first recorded adoption in a particular country. Figure 1.4 plots the average adoption lag against the year of invention for a large set of major technologies introduced over the past 250 years. It draws on a historical dataset originally assembled by Diego Comin and Bart Hobijn, which was extended to include more recent technologies for the purposes of this report.
The pattern is striking: the newer the technology, the shorter the lag before it reaches other parts of the world. In other words, the time between invention and first use has drastically shortened.
Consider a few examples. The telegraph, invented in the first half of the 19th century, took on average almost 50 years to reach countries around the world. The automobile, emerging in the late 19th century, diffused faster, with an average lag of about 36 years. By contrast, cellphones, introduced in the 1970s, saw first use globally within less than 20 years. More recent generations of mobile technologies – 3G and 4G – have spread even faster, often reaching new markets within just a few years of their introduction.
At the most extreme end of this trend are large language models (LLMs), exemplified by the release of ChatGPT in November 2022. Within days of becoming available online, users in virtually every country had accessed and experimented with the technology (see Box 1.1).
Artificial intelligence and machine learning have been part of the technological landscape for decades. But the rise of GenAI – beginning in 2022 with tools such as ChatGPT, Claude, Gemini, Copilot, and Perplexity – has been exceptionally fast and transformative. Beyond producing text, code, images, or other content, GenAI can deliver a wide spectrum of knowledge tailored to users’ individual questions and needs, with a wide array of practical applications.
Since late 2022, GenAI services have spread across the world at unprecedented speed. Early usage was heavily concentrated in the United States – more than 70 percent of global traffic at launch – but this dominance faded rapidly. Within one month, the US share had fallen to around 25 percent, and later stabilized near 20 percent, as traffic spread quickly to a wide range of economies. By mid-2023, ChatGPT alone was attracting roughly 500 million unique users a month – equivalent to about 12.5 percent of the global workforce – underscoring the breadth of early international uptake.
Across countries, there is a clear positive correlation between income levels and GenAI use (see Figure 1.5). Higher-income economies tend to show greater overall activity, reflecting differences in digital infrastructure, connectivity and skills
At the same time, many low-income economies remain in the early stages of adoption due to constraints such as limited internet access, insufficient data-center capacity, and shortages of digital and AI skills. Language coverage also shapes adoption patterns, as many GenAI systems perform strongest in English.
GenAI’s rapid global spread is facilitated by the fact that it runs on existing digital devices and that many tools are available at low or no cost – reflecting intense competition among providers seeking to establish an early market foothold.
Not every technology fits this trend. Electric vehicles (EVs) stand out as a relatively new innovation with a still-longer adoption lag. Unlike digital technologies, EVs depend on extensive physical infrastructure – such as charging networks and grid capacity – that takes time and investment to build. In addition, policy incentives encouraging EV adoption have only emerged more recently, further explaining their delayed take-off.
Even so, the overall trend is unmistakable: new technologies are reaching more places faster than ever before. The world is catching up more quickly – and the distance between invention and first use is steadily shrinking.
The adoption lag data also reveals persistent patterns of technological leadership and followership across regions. Advanced economies consistently emerge as early adopters, typically embracing new technologies 20–80 years before the global average, while Africa shows the opposite pattern, adopting most technologies 10–50+ years after the global average, while Asia and Latin America show mixed patterns depending on the specific technology (see Figure 1.6).
These differences matter greatly for global development. Countries that adopted technologies such as electricity, railways or fixed-line telecommunications much earlier than others accumulated decades of productivity gains – advantages that still shape today’s global income gaps. Yet the data also points to new possibilities for catching up. More recent technologies, like the internet and mobile phones, have spread far more quickly across borders. This has made it easier for many developing economies to adopt new technologies sooner and, in some cases, leapfrogging older ones in the process.
Stylized facts #4: A widening use gap
Adoption lags tell us when a country first starts using a new technology, but they do not show how quickly that technology spreads within the country once it has arrived. To understand this second part of the story, we need to look beyond the moment of first adoption and examine how widely a technology is taken up overtime.
Based on the historical database described above and following the methodology of Comin and Mestieri,
Figure 1.7 illustrates how the usage gap between advanced economies and developing economies has changed over time. From the 1800s through much of the 20th century, this gap generally widened. Although the pattern varies across individual technologies, newer technologies typically saw larger differences in how intensively they were used. Comin and Mestieri
The picture shifts, however, once we include more recent technologies such as 3G, 4G, and wind power. For these innovations, the intensity of use has begun to converge across countries. This is an encouraging development, suggesting that today’s digital and renewable technologies may offer greater chances for developing economies to narrow historical gaps.
Figure 1.8 provides a complementary perspective by comparing technology-use gaps across developing-country regions relative to advanced economies. With some exceptions, Africa exhibits the widest gaps across most technologies, followed by Latin America and then Asia. Importantly, while all three regions show declining gaps for the most recent generations of technologies. Asia stands out: not only has it narrowed its gaps more substantially, but in some cases even displays a technology-use edge – that is, usage levels that exceed those observed in advanced economies.
How does GenAI fit into this picture? Because of its very recent emergence, GenAI is not included in the analysis underlying Figures 1.4 and 1.6–1.8, as we do not yet have the long-run data needed to estimate its potential use intensity in the same way as for earlier technologies. Nonetheless, initial indications point to a rapidly rising use intensity – particularly in several middle-income economies that appear to be adopting GenAI at levels well above what their income would predict (see Box 1.1). Much like 3G and 4G, GenAI relies heavily on preexisting digital infrastructure, suggesting that its diffusion trajectory may also show signs of convergence with advanced economies over time.
These findings need to be interpreted with care. Some of the narrowing may simply reflect differences in the characteristics of newer technologies – for example, their lower costs, greater adaptability, and ease of scaling across markets. Moreover, the set of technologies covered in this analysis is not exhaustive. Even so, given the central role of digital technologies as general-purpose technologies, the emerging signs of convergence in use intensity are a positive signal for global development.
What determines faster and wider technology diffusion?
The pace and breadth of technology diffusion vary greatly across innovations, industries and countries. While some technologies spread globally within a few years, others take decades to become widely used. Understanding what drives these differences is crucial for designing policies that accelerate beneficial diffusion. Four broad sets of factors stand out in the economic literature.
The nature of technology itself
Some technologies spread rapidly either because they are simple to use, inexpensive or immediately beneficial to a wide range of users. Others diffuse more slowly because they are costly, complex or require complementary investments. A key determinant is the price of the technology relative to the benefits it offers. Technologies that provide clear value at a manageable cost tend to spread more quickly.
Another important determinant is the need for supporting infrastructure. Some innovations depend on large and costly networks – such as electricity grids, broadband cables or charging stations – that take time to develop. Others are able to leverage existing infrastructures and therefore diffuse more easily. As pointed out above, the rapid global spread of LLMs reflects the fact that the underlying digital infrastructure – the internet – was already in place worldwide. In contrast, technologies like EVs or renewable energy systems require heavy new investments in physical networks, such as charging stations, transmission grids and energy-storage facilities. Such infrastructure requirements can significantly delay adoption, even when the technologies themselves are mature and cost competitive.
The speed of information
Diffusion depends not only on the intrinsic appeal of a technology, but also on how quickly and widely information about it circulates. In the 19th century, knowledge about new inventions traveled only as fast as traditional mail and newspapers could carry it. The arrival of the telegraph, followed by the telephone, radio and fax, progressively accelerated the exchange of technical knowledge. The internet further transformed information flows, and today, digital platforms – and increasingly, AI-based systems – allow near-instant access to vast repositories of technical and scientific information.
A particularly striking recent development is the emergence of LLMs. Beyond simply granting access to large quantities of information, LLMs can process and apply that information in highly accessible ways, helping users generate or retrieve highly specific, context-relevant knowledge on demand. This capability reduces the cost and effort of finding, interpreting and adapting information, thereby lowering barriers to learning and potentially accelerating technology diffusion itself.
These advances have greatly reduced the time it takes for new ideas to reach potential adopters around the world.
Absorptive capacity and local capabilities
Even when information is freely available, not all users can immediately make use of a new technology. Adoption often requires absorptive capacity, the ability to understand, adapt and apply new knowledge effectively.
Building absorptive capacity depends on education, technical training, research institutions, and linkages with global knowledge networks. As the next chapter will explore in more depth, these innovation capabilities are decisive for how well economies learn from others and close the gap between invention and effective use.
Public policy and institutions
Public policy plays a pivotal role in shaping technology diffusion. Governments influence diffusion through the following multiple channels.
Complementary public infrastructure
Technology adoption by individuals or firms often depends on the presence of public goods such as roads, ports, electricity, and telecommunications networks. The success of mass-produced automobiles, for instance, relied on publicly-funded road systems. Similarly, the diffusion of digital technologies depends on the availability of affordable broadband and mobile coverage – key areas where policy intervention can accelerate adoption.
Standardization and interoperability
Standards ensure that products and systems work together, reducing uncertainty for producers and users alike. In many cases, standards emerge organically from industry collaboration. Private or industry-led standards – such as those developed by professional associations or technology consortia – often prove highly effective in promoting interoperability and market expansion. However, in other cases, government involvement in standard-setting is essential. Public standards can ensure safety, compatibility and consumer protection within sectors where coordination failures or high risks might otherwise slow adoption. Examples include railway gauges, electrical voltage, control of the radio spectrum, building codes, and automotive emission norms. Governments and regulatory authorities often set or coordinate such standards directly, sometimes through international bodies like the International Organization for Standardization (ISO) or the International Telecommunication Union (ITU).
Safety and consumer protection
Regulations that guarantee safety and quality can also foster technology diffusion by building public trust. Vehicle safety inspections, aviation certification and pharmaceutical regulatory approvals ensure that new products meet minimum safety thresholds before entering the market. This helps consumers overcome skepticism toward unfamiliar technologies, enabling broader uptake.
Intellectual property (IP) policy
IP systems influence how quickly and widely technologies spread by balancing a fundamental trade-off between incentives for innovation and access to new technologies. Exclusive rights encourage firms and inventors to invest in developing new ideas, while disclosure requirements and time limits ensure that others can learn from them and build on their advances. Published patent documents, in particular, form a vast global repository of technical information that supports imitation, learning and further innovation.
IP rights also help turn inventions into tradable assets. By defining ownership and offering legal assurance, they make it easier for firms to license, buy or sell technologies, thus enabling markets for knowledge and facilitating cross-border diffusion. At the same time, as discussed above, adopting technologies invented elsewhere often requires adapting them to local conditions. This follow-on innovation is supported by a range of IP instruments – including utility models, design rights, and trademarks – which strengthen incentives for incremental and adaptive innovations that help technologies take root within diverse settings.
Conclusion
Technology diffusion is a broad and multifaceted process at the core of economic development and human progress. For technological breakthroughs to drive economic growth, they must spread widely throughout an economy. While the pace of global technology adoption has accelerated dramatically, the benefits of new technologies remain unevenly distributed both within and across countries.
Several key insights emerge from this introductory chapter. First, diffusion is never automatic; even superior technologies face obstacles that can delay or prevent their spread, from infrastructure requirements and skill gaps to regulatory barriers and financing constraints. Second, the nature, cost and infrastructure requirements of a technology fundamentally shape its diffusion pattern. Third, the flow of information and the presence of local absorptive capacity determine whether potential adopters can effectively learn about and implement new technologies. Finally, public policy and institutional frameworks play a decisive role in creating the conditions required for successful diffusion, as the spread of technology does not of itself automatically translate into desired outcomes in economic, social and environmental terms.
These insights have important implications for policymakers seeking to harness technological progress for economic development and social benefit. Solely encouraging invention is insufficient. Equal attention must be paid to creating conditions that enable rapid and broad diffusion. This includes investing in complementary infrastructure, building human capital and institutional capabilities, establishing appropriate regulatory frameworks, and designing IP systems that balance innovation incentives against technology access.
The patterns of technology use between advanced and developing economies, as highlighted in this chapter, not only emphasise the source of persistent global income gaps, but also point to the enormous potential for catch-up growth if barriers to diffusion can be reduced. Meeting today’s pressing global challenges – including climate change – will depend on countries’ ability to diffuse beneficial technologies rapidly and widely.
Achieving this will require close cooperation and interdependence among the many actors that make up the global innovation ecosystem. The chapters that follow build on this foundation by examining specific mechanisms through which technological knowledge spreads globally and exploring detailed case studies that illuminate these concepts in practice. Together, they aim to provide actionable insights for policymakers, business leaders, and researchers working to unlock technology's transformative potential for societal benefit.