Newsletter
Development and Application within Artificial Intelligence Industry
The history of artificial Intelligence (AI) can be traced back to the mid-twentieth century. After a long period of low development, advances in hardware (e.g., computing chips) and algorithms (e.g., neural networks) allowed AI to flourish again in the late 20th century.1 AI covers many core technologies, including hardware devices such as chips, cloud computing systems and terminal devices, as well as software algorithms such as artificial neural networks, deep learning, image recognition, and big data trend analysis, etc. AI is widely applied in a broad range of fields, including Fintech, manufacturing, healthcare, transportation, law, agriculture and defense. Generally speaking, AI applications may include: applications in academic research, the creation of new digital intelligence services, and the introduction of AI technologies to existing industries.
1. Application of AI
In academic applications, data (e.g., astronomical satellite images) that had previously taken scientists decades to collect can now be collected in a short period of time with the help of powerful graphics processing units (GPU) and deep learning. 2 Meanwhile, AI can also improve natural disaster warning systems, making them faster and more accurate, thereby reducing casualties and damage to property. Furthermore, in response to the rampant coronavirus pandemic, AI can also be applied to research on inhibiting the spread of infectious diseases, to help researchers quickly and accurately deconstruct viral proteins, track and predict the trend of viral mutations, and speed up vaccine development. 3
With respect to the creation of new digital intelligence services, people-centric4 digital intelligence services are constantly evolving. The application of AI-based image recognition in shopping malls and department stores, for instance, not only allows understanding of the distribution of customer traffic and promotion of advertisements that match consumers' preferences via real-time push notifications, it also reduces waiting time. In medical applications, AI enhances and facilitates the ultimate goals of preventive medicine, such as telemedicine, remote patient monitoring, integrated electronic medical record management systems, wearable devices for patients, online medical consultations, registration appointments, smart long-term care applications, or AI smart outpatient clinics.5
As to the introduction of AI technologies to the existing industries, for instance, the combination of the Internet of Things (IOT), Edge Computing and AI technologies enables connectivity among working areas or machines in factories. Hence, automated machines can send data to terminal systems for integration and autonomous analysis. Subsequently, feedback on decision instruction is automatically sent back to each working area, machine or monitoring team. In this way, enormous human resources can be saved, and the decrease in productivity caused by machine abnormalities can be avoided while the value of production is increased. 6
2. Prospects of AI and the Patent Portfolio
As technology and manufacturing industries prevail in Taiwan, Taiwan's Ministry of Science and Technology regards AI as the most important technology for future industrial development, and continues to promote related projects such as the “Semiconductor Moonshot” program and the “Formosa Grand Challenges,” 7 in regard to the importance to AI application in these industries. Moreover, in mainland China, AI was first included in the State Council's “Work Report of the Government” 8 in 2017, in which AI officially became one of China's national strategies. Since then, focus has shifted from basic AI research to the integration of technologies and industries. Driven by policies, commercial applications of AI (in terms of the number of enterprises, investment scale, market scale and the like) in mainland China have ranked among the top in the world. The deep integration of AI with many industries will impact the traditional business process and industrial chain, and further facilitate the formation of industrial intelligent innovative ecosystems.
There is still considerable room for the development of AI technologies in various fields. Although the AI technologies used in different fields may have similar or equivalent contents and concepts, if the particular technical characteristics of the AI device or technology can be manifested and can bring unanticipated technical effects, then it is possible to protect the core technology through patent rights.9 Nowadays, AI is needed in every field for greater benefits. Even though ethical issues in AI are still of great concern, it is undeniable that the future of human life will be changed by the evolution of AI technology.
9鄭光益,人工智慧類發明的說明書及申請專利範圍撰寫,https://www.taie.com.tw/tc/p4-publications-detail.asp?article_code=03&article_classify_sn=64&sn=1370 (2019年9月)。