In the 2018 edition of its Science, Technology and Innovation Outlook report, the OECD examined trends, new policies and instruments being used in science, technology and innovation policy in OECD member countries. A key discussion within the report was the interconnected relationship between digital technologies, such as AI, and public funding.

This is because as digitization becomes more commonplace and vital across numerous disciplines and markets, more data is being collected and in need of analysis. To address these analysis concerns, new digital technologies are required; to address the R&D investments required for innovating these new digital technologies, additional funding—including federal funding—is required; to optimize investments for digital technologies, governments and public funding institutions require a solid comprehension of how these new digital technologies work by analyzing their functions and data collections; and to analyze those data collections, these public institutions also require their own digital technologies that can handle the vast amounts of data that technologies such as AI can gather.

Thus, digitization is not only a major aspect of future innovation across industries and the improvement of economies around the world, but also a central driver of the future of funding instruments and mechanisms. In its report, the OECD examined the wide impact of digital technologies such as AI as well as the efficacy of federal funding instruments, and how they can be improved to be better prepared for what the Organization calls the “next production revolution.”

The AI Factor

The growing prevalence of AI is serving as a catalyst for many industries, and, as the OECD stated in its report, AI could establish entirely new industries altogether. Moreover, technologies that depend on AI for their advancement, such as biotechnology or 3D printing, have great potential for accelerating economic growth. As the report highlighted, the development and implementation of new production technologies is imperative, not only for the future of economic productivity across numerous industries, but also for raising the standards of living and labor productivity growth in some OECD member countries.

Digital technologies are able to ramp up productivity through numerous methods, such as reducing machine downtime through prediction techniques, increasing precision and efficiency, and developing green technologies that benefit the environment (i.e., nanotechnological materials that have the capability to cool themselves down without consuming energy).

AI technology is in an almost-constant state of evolution, making it imperative that pubic institutions gain a better understanding of rapidly changing digital technologies.  

AI has already helped drive growth in numerous industries. For example, within pharmaceutical applications, AI is said to become the “primary drug discovery tool” by 2027, according to an industry expert from AstraZeneca, cited in the report. Thanks to AI’s support in managing bioinformatics, analyzing genomic and drug safety data and improving in silico modeling, AI has more than proved its utility within the industry. AI is also useful in oil applications, as it has the capabilities to examine and analyze photographic imagery of the interior of oil pipelines, searching for microscopic fissures.

In the mining industry, AI is effective in the exploration for mineral deposits and the operation of autonomous drills, ore sorters, loaders and haulage trucks. AI is also the key driver of analyzing massive volumes of IoT data, as it has the ability to process combined sensory data and text. Because of this, the OECD forecasted that AI’s impact on future production could drive the creation of completely new industries, comparing it to the new industries that were spurred by the discovery of DNA structure in the 1950s.

The report explained that since many companies are in possession of large, valuable datasets that they may not be using effectively, it would behoove them to partner with AI startups and other businesses using AI to help determine the datasets’ value and the best way to utilize it. The governments, the OECD stated, can serve as “honest brokers for data partnerships,” working alongside stakeholders to establish agreements for data sharing amongst companies. Creating more open-data initiatives are also a way to ensure the use of data for the public benefit, according to the report. However, the OECD indicated that it is essential that regulations on AI and the impacts of those regulations be monitored carefully and regulatory reviews be common—this is due to the fact that AI technology is in an almost-constant state of evolution, making it imperative that pubic institutions gain a better understanding of rapidly changing digital technologies.

Public Research Funding Trends

Governments also play a significant role in the funding and growth of scientific research and, therefore, innovation for technologies such as AI. Therefore, a key issue raised by the OECD was the the efficacy of funding instruments, specifically, what the merits of accomplishing specific policy goals are. In examining this, the OECD examined several country reviews and evaluations of programs that support research around the world to determine how particular funding instruments are adapted to certain policy goals, as well as how these funding processes are implemented at Higher Education Institutions (HEI) and Public Research Institutes (PRI).

While the 2008 recession negatively impacted R&D funding, research continues to be and will likely remain a significant aspect of government budgets, especially as trends such as digitization and centralized databanks continue to rise.

Scientific progress is the main driver of innovation, and it is especially important, the OECD states, at knowledge institutes like HEIs and PRIs. However, in 2016, HEIs and PRIs represented approximately 18% and 11%, respectively, of Gross Domestic Expenditure on R&D (GERD) in OECD member countries. This figure is much lower than the business sector, which made up 69% of 2016 GERD in OECD countries, yet HEIs and PRIs conduct over 75% of total basic research. HEIs especially are important in the fostering of R&D, while the influence of PRIs has diminished in certain countries. HEIs are vital to scientific research because not only do they provide higher education opportunities, but they are also deeply involved in what the OECD called “longer-term and higher-risk scientific knowledge,” and have an indelible influence on applied research, knowledge and technology transfer, and other innovation endeavors.

While there are variations amongst countries, federal governments are the main sources of academic research. In 2015, government allocations propelled 67% of HEI research and 92% of PRI research. While the 2008 recession negatively impacted R&D funding, research continues to be and will likely remain a significant aspect of government budgets, especially as trends such as digitization and centralized databanks continue to rise.

The Evolution of Publicly Funded R&D

While the portion of government funding for PRIs has been generally stable since the 1980s, it has been on a steady decline for HEIs, which have been increasingly obtaining third party funding. While federal HEI R&D funding had seen a revival a couple years after the recession, by 2010–2011, the increases to R&D budgets had reverted or slowed down. GERD to HEIs and PRIs decreased, even though economic growth continued.

Research project information necessary to determine the proper channels of funding allocations can be directly retrieved through data processing, reducing the need and costs for expensive award competitions.

The OECD argued that this continuing decrease cannot be blamed on budgetary pressures, but is actually a corroboration of anecdotal evidence from experts in the science industry that suggest a “frustration” regarding the lack of enough tangible, real life innovation results from past GERD allocations to R&D. This also creates hurdles for science advocates when they are negotiating with ministries and representatives that are responsible for science-related policy areas.

The OECD separated publicly funded R&D into two categories:

• Competitive project funding: in this mechanism, funding instruments, agencies, councils or ministries provide resources for a particular research project, one that has specific scope, budget and time requirements. This mechanism allows researchers to apply for funding, and financial awards vary in their size, time period and winner (i.e., an individual versus a research center)
• Non-competitive institutional funding: this funding mechanism provides a block of funding in more general terms, without specifying particular R&D projects or programs. This type of funding is usually provided as an annual federal allocation to HEIs or PRIs, not specific research groups, and the funds are used not only for innovation, but also for administrative purposes, such as staff wages, infrastructure, and maintenance and upgrades that are related to R&D. Institutional funding used to be for specific research projects, but is now typically allocated as a block grant, or a lump sum, giving institutions more freedom in how they want to spend the funds.

The OECD stated the evolution of funding has blurred the lines between competitive and non-competitive funding instruments, and that this has led to the need for more “nuanced” frameworks for research funding. The OECD proposed that the separation between competitive and non-competitive funding requires a revamping into a single analytical framework, and suggests that certain factors—namely, competition, granularity, competition level, selection/allocation criteria and orientation/directionality—should be taken into consideration when creating this new funding framework.

Digitization is cited by the OECD as a major gamechanger, not only within the science industry, but also in regards to funding instruments. As digital technologies such as AI continue to transform the R&D landscape, they are also providing policymakers and funding mechanisms with the capability to monitor research. This is because more current research has been made available, thus allowing a more thorough analysis and evaluation through digital innovations. Research project information that is necessary to determine the proper channels of funding allocations can be directly retrieved through data processing, which greatly reduces the need and costs for expensive award competitions. As digitization and AI continues to grow, governments are looking more towards cost efficiency and accountability within scientific R&D when considering budgets, and it will be imperative to design new funding instruments and programs to address these concerns.