What are the impacts of artificial intelligence on manufacturing and the economy?

The San Francisco Federal Reserve and the Federal Reserve System Innovation Office are working closely together to recently launch the Emerging Technology Economy Research Network, a new initiative aimed at better understanding how emerging technologies such as artificial intelligence are shaping the future economy. Since the 18th century, the Industrial Revolution has triggered a series of reforms that have completely changed the face of the entire society. We may be in the early stages of a new technology era, the era of artificial intelligence, which could trigger a similar scale of change.

Firstly, artificial intelligence experiments are extensive and accelerated. The application scope of artificial intelligence in specific businesses and respective industries. The report includes extensive experiments on artificial intelligence tools, including GenAI, while pointing out that the actual integration of artificial intelligence with established operations is more limited. I often try GenAI in office operations, such as generating marketing copy, manufacturing schedules, and product labels. This technology is particularly useful for completing tedious large-scale tasks, such as writing product descriptions.

In addition, as companies shift from experimenting with artificial intelligence to deploying tools as permanent operational features, they are combining supplier software with embedded artificial intelligence tools and publicly available large-scale language modeling tools. Some reports suggest creating new positions for experts who have received training in using artificial intelligence tools, although the current demand for such skilled positions exceeds the number of available technicians and engineers.

However, artificial intelligence has limitations in the manufacturing industry. Accelerated testing of GenAI tools in certain areas of business operations, but also emphasized some true limitations of using artificial intelligence tools in the manufacturing process. These limitations are particularly applicable to the physical aspects of advanced manufacturing, such as precision machining of parts. The main reason for this situation is that there is currently not enough data to train the use of GenAI tools in the manufacturing process. Due to the fact that companies must be careful to protect proprietary information from being used to train third-party and public GenAI tools, developing sufficient training data becomes even more difficult. Advanced manufacturing companies must meet strict quality and personal safety standards. The existing GenAI technology is not yet able to consistently meet these standards, posing reputational, physical, and economic risks to any company deploying these technologies. This in turn leads manufacturing companies to focus their implementation efforts on areas with lower value and lower risk. Due to cost considerations, the integration of GenAI and advanced manufacturing will lag behind. The machines used in this industry are very expensive and have a long service life. Before purchasing new artificial intelligence machinery, manufacturers will first ensure that their early investments can be rewarded by using existing machines at the end of their lifecycle.

Obviously, high skilled professions can also expect to benefit from the complementary benefits provided by artificial intelligence to the greatest extent possible. Radiologists use this technique to enhance their ability to analyze medical images. For these reasons, as developed economies have a larger proportion of affected workers, but only a small portion may maximize productivity, the overall impact on these economies may become more polarized.

On the other hand, economically underdeveloped countries are less affected by artificial intelligence. In emerging markets such as India, where agriculture dominates, less than 30% of employment opportunities are affected by artificial intelligence. In Brazil and South Africa, this number is close to 40%. In these countries, the urgent risks brought by artificial intelligence may be relatively small, but the opportunities for improving productivity may also be limited.

In summary, over time, labor-saving artificial intelligence may threaten developing economies that heavily rely on labor-intensive sectors, especially in the service sector. Tasks outsourced to emerging markets may return to developed economies and be replaced by artificial intelligence. This may put the traditional competitive advantage of developing economies in the global market at risk, and may make income convergence between them and developed economies more difficult.