AI Patent Acquisition Strategies

Pine IP Firm
December 24, 2024

AI technology is expanding its impact across the economy and society, extending beyond fields such as healthcare, bio, finance, manufacturing, autonomous driving, robotics, and the Internet of Things (IoT), enabling companies and research institutions to create new added value. Consequently, the number of applications for AI-related technologies is rapidly increasing in the global patent market. Major patent offices worldwide (EPO, JPO, KIPO, CNIPA, USPTO) are continuously striving to assist applicants by reorganizing their examination standards for AI-related inventions and providing clear guidelines. This column, based on data from the IP5 patent offices, explores strategic approaches and key points for securing artificial intelligence patents and proposes methods for building a differentiated patent portfolio utilizing these insights.

Understanding Global AI Patent Trends and Analyzing National Examination Standards

In an environment where AI technologies are interconnected and cross-utilized globally, patent strategies solely targeting the domestic market may face limitations. The IP5 patent offices (European Patent Office EPO, Japan Patent Office JPO, Korean Intellectual Property Office KIPO, China National Intellectual Property Administration CNIPA, United States Patent and Trademark Office USPTO) systematically disclose AI invention examination guidelines, case law, and case studies, allowing applicants to understand and prepare for subtle differences in national examination standards.

Patent Eligibility: Emphasizing Technical Features and Eliminating Abstraction

The first consideration when filing an AI technology patent application is the requirement for patent eligibility. Mere mathematical algorithms or abstract concepts may be excluded from patent protection. To prevent this, the invention must possess concrete and substantial technical features that solve real-world technical problems.

  • EPO: Within the guidelines for examining computer-implemented inventions (CII), we determine whether there is a specific technical contribution for AI-related inventions.
  • USPTO: Based on 35 U.S.C. ยง101, we emphasize the tendency to grant patent eligibility to inventions with technical improvements beyond 'abstract ideas.'
  • JPO, KIPO, CNIPA also examine patent eligibility by comprehensively reviewing implementation methods, data processing procedures, and the interaction between hardware and software.

Based on this, applicants must reduce the degree of abstraction and secure patent eligibility by clearly describing specific technical features, problem-solving methods, and technical effects in the specification and claims.

Meeting Specification Requirements: Ensuring Clarity, Enablement, and Support

AI technology involves complex elements such as algorithms, model structures, datasets, and training and inference processes. Therefore, strategic descriptions are necessary to meet the requirements of clarity, enablement, and written description (support).

  • Detailing Specific Implementations: It is necessary to specifically describe the AI model's training process, types of data applied, pre-processing and post-processing steps, and hardware requirements to prove that it is a feasible technology, not just a simple idea.
  • Utilizing Case Examples: By referring to AI-related specification examples provided by KIPO, JPO, EPO, etc., one can understand the level of technical detail required for examiners to have sufficient understanding. This can prevent rejections due to ambiguity, non-enablement, or lack of support.

Strategies for Securing Novelty and Inventive Step

In the strategy for securing AI technology patents, the assessment of novelty and inventive step is a key point. Simply applying AI algorithms to existing technologies, such as slightly improving performance using open-source libraries, makes it difficult to obtain recognition for inventive step.

  • Presenting Clear Technical Differentiation: Emphasize distinct differences compared to existing technologies, such as new data processing methods, maximization of learning efficiency, solutions to specific industrial problems (e.g., improving medical diagnostic accuracy, automating manufacturing processes), or innovations in hardware structure.
  • EPO's COMVIK Approach: The method of distinguishing technical and non-technical features in an invention and judging inventive step based on solving technical problems can also be applied to AI inventions.
  • Utilizing Case Studies: Understanding which elements are recognized for inventive step based on comparative studies by EPO, JPO, KIPO, CNIPA, or cases from the USPTO can be of great help in establishing an application strategy.

Utilizing Comparative Studies and Joint Seminar Materials by Country

The IP5 patent offices support applicants in understanding different examination standards through international comparative studies of examination guidelines and standards for AI-related inventions, joint symposia, and seminars. Utilizing these comparative materials allows for early identification of country-specific standards to formulate application strategies or to concretize strategies for securing patent protection in one market and then expanding to others. In particular, by comparing and analyzing similarities and differences between countries through comparative research results between JPO, EPO, and KIPO, or through materials from international symposia, more efficient global patent strategies can be developed.

Developing Customized Strategies for Each AI Technology Field

AI is not merely 'a technology' but a convergence technology that drives innovation by combining with various industrial fields. For example, the requirements and technical challenges differ for fields such as medical AI, manufacturing AI, autonomous driving AI, and fintech AI, and patent offices in each country may provide examination guidelines that reflect these specificities. Therefore, applicants must analyze their respective technology fields and closely examine cases and examination trends for AI inventions in those fields. This enables the implementation of field-specific claim configurations, specification description strategies, and pilot applications in various countries, ultimately leading to a qualitative and quantitative expansion of the patent portfolio.

Continuous Portfolio Management and Response to Competitive Environments

AI technology is rapidly evolving, and it is difficult to secure sustained competitive advantage through one-off patent applications.

  • Building a Hierarchical Patent Portfolio: After securing core patents, a defensive and offensive patent network can be formed through subsequent applications for improvement inventions, application inventions, and modified algorithms and data processing methods surrounding them.
  • Monitoring Standardization Trends: If AI-related standards are established, active monitoring of standardization trends and strategic applications may be necessary to secure Standard Essential Patents (SEPs).
  • Licensing and Collaboration Strategies: The AI ecosystem is complex, so IP strategies through patent license transactions, technology collaborations, and establishment of joint ventures between companies should also be considered.

Utilizing Expert Personnel and Institutions

Finally, to gain insight into the differing examination standards by country and complex technical content, collaboration with specialized patent attorneys or patent law firms is essential. Experts can assist in developing customized strategies by considering trends at each patent office, case law, and industry best practices. Furthermore, by utilizing patent analysis tools and AI-based prior art search tools, comprehensive patent strategies can be established, including understanding competitor trends, identifying white spaces, and analyzing potential infringement risks.

Conclusion

The strategy for securing artificial intelligence patents goes beyond simply protecting inventions; it plays a crucial role in securing a company's future technological leadership, establishing market preeminence, and fostering long-term business growth. By fully utilizing the guidelines, case law, and comparative research results provided by the IP5 patent offices, one can grasp national examination standards and trends, and develop effective strategies for specification description, claim configuration, and securing novelty and inventive step. Furthermore, through field-specific strategies, portfolio expansion, and the utilization of expert personnel, a solid patent position can be established in domestic and international markets. Such comprehensive and systematic AI patent acquisition strategies will strengthen competitiveness in the AI innovation ecosystem and contribute to the long-term success of businesses.

If you wish to reliably protect your innovative ideas in the global market, it is essential to develop a strategy with professional patent experts. Pine IP Firm, with its extensive experience in advanced fields such as AI and ICT, supports the construction of customized patent portfolios and the efficient acquisition of rights.