The semiconductor industry is in the midst of a global race to achieve sub-1nm processes, with innovations in transistor design as the key driver. As patent filing continues to develop across multiple design technologies such as commonly used Fin Filed Effect Transistors (FinFET) and the new generation of Gate-All-Around FET (GAAFET), how does data across the 14,000 patents linked to these technologies give us insight into the state of bleeding-edge transistor technology, who the leading players are and where the opportunities and challenges lay for both incumbents and challengers in the space?

This study, done in collaboration between Global Prior Art (GPA), a world leading IP analysis and strategy consulting firm, and IPRally, provides a first of its kind analysis of the next generation of key transistor design technology through the detailed lens of patent holdings by major players in the industry. By combining expertise in semiconductors with a powerful, patent-trained AI tool, designed for analysis of technology and patents, this report yields unprecedented data, analysis and insight into this complex space, highlighting the key to creating fact based proactive strategies at the corporate and country level when undertaking in-depth, strategic patent analysis.

‍

Data & Methodology

Identifying who holds the keys to the next generation of semiconductor technology is challenging due to the complexity and size of the underlying patent space, which comprises more than 5,000 GAA, and 9,000 FINFET patent families* covering the current leading-edge IC technology in the U.S. alone. Patents provide critical insight into future trends, given their level of technical detail, thus they allow a deep understanding of technical innovation and competitive strategy that is not available with high-level market analysis.

(* A patent family refers to a group of related patent applications or granted patents that cover essentially the same invention within and across various jurisdictions.)

‍

To gain insight into this complex space, a transformative methodology was employed which combined the technical expertise of Global Prior Art’s (GPA) team of IC experts who reviewed relevant patents and IPRally’s powerful AI Classifier trained to identify relevant patents hidden among the tens of thousands of semiconductor patents.

GPA’s research into the patent filings in this space found that until recently important patents filed worldwide relating to semiconductor technology typically included a U.S. patent publication, including patent documents originally filed in Europe and Asia (such as Korea, China, Taiwan and Japan). This was typically the case due to the critical importance of the large U.S. market. 

However, this is no longer true.

The Rise of the Chinese Semiconductor Industry

A deeper analysis, utilising IPRally’s machine translation of foreign language patents, found that Chinese players have filed a large number of patents in the last few years only in China, covering advanced semiconductor technology. This change appears to reflect a focus on the large domestic market, as well as a national priority to create an advanced semiconductor sector. 

This filing approach has resulted in a false impression in the U.S. and Europe that Chinese IC technology is 5+ years behind that in the U.S. To avoid being blindsided by technology development in China, or having important patents challenged by published Chinese applications, it is essential to follow the technology and IP activity in China and its implications. 

To facilitate this, the study’s coverage was extended to also include Chinese language activity.

Using the IPRally Classifier

The study is powered by human intelligence and based on the expertise of GPA's full-time technology specialists with decades of experience at the intersection of IP, technology strategy, and the semiconductors field. 

Their custom analysis and insights take into consideration the following key patent quality/value characteristics when shaping our selection of subcategories and taxonomy creation:

• Impact, timing, and novelty of claimed technology,
• Breadth of claims,
• Discoverability and likelihood of infringement.

Leveraging subject matter expertise plus IPRally’s AI classifier, which was specifically designed to analyze patents and the underlying technology, allows insight into a vastly complex field that can be found nowhere else. 

Manually identified and hand-curated patents were used to train IPRally’s AI classifier and help expand the coverage with the resultant patents then manually reviewed to confirm if they should be included in the training data. The use of curated training data alongside expert analysis and review allowed for high-confidence classification to be conducted across thousands of patent families.

This hybrid approach yields the best of both worlds: Expert human insight into a rapidly evolving and highly complex innovation field, paired with proprietary, patent-specific AI classification to analyse large, complex datasets. IPRally’s custom approach allows for a user-specific taxonomy, enabling classification independent of traditional patent classification systems and tailored to the user's needs and technical perspective.

‍

Key Findings

Overlooked Chinese progress

Contrary to common beliefs, China has made notable gains in advanced semiconductor development in recent years. 80% of Chinese next-gen transistor patents are filed only in China, posing a risk to non-Chinese semiconductor players.

Rapid technology transition

The shift from FinFET to GAAFET is occurring at a rapid pace and new GAAFET filings have already overtaken FinFET, with yearly patent family filings for GAAFET over 3 times that of FinFET in 2022.

Current top U.S. semiconductor technology filers

TSMC, Samsung, and IBM are the leading filers of FinFET and GAAFET patents in the U.S., collectively holding about 50% of all patents in this segment.

Familiar technology leaders

Samsung dominates U.S. patent filings in FinFET while TSMC is tied with Samsung in the emerging GAAFET technology.

AI’s emergence in supporting complex patent analysis

Combining expertise in semiconductors with a powerful AI tool, designed for analysis of technology and patents, yields unprecedented data, analysis and insight into this complex space, highlighting the key to creating fact based proactive strategies at the corporate and country level undertaking in-depth, strategic patent analysis.

Read the full report

To read the full report and analysis click the link below to access the full PDF version.

‍

Sign up for our upcoming Classifier webinar

Join us on February 13th at 5pm CET (11am ET) for a deeper dive into AI patent classification with our research partners, Global Prior Art.

In this webinar, we will dive into the methodology behind the research and showcase how custom AI classification supported the analysis as well as showcasing the different benefits and use cases of AI classifers. Sign up using the link below.

       

‍

About Global Prior Art

Global Prior Art (GPA) is a pioneer and world leader operating at the intersection of intellectual property (IP),  technology analysis and strategy. With over 40 years of experience in the microelectronics and semiconductors space GPA’s clients include many leading companies, VCs, and law firms.

GPA’s staff consists of engineers with advanced degrees in electrical engineering, materials science, optical engineering, software and telecommunications from leading schools.  Its deep technology expertise, and experience addressing 20,000+ matters ensures unrivaled effectiveness. Headquartered in Boston, GPA supports leading companies and top law firms in the U.S. and abroad.   Clients rely on our expertise to guide decision making, in high-stakes matters involving IP, including  high stakes litigation, vetting patents for acquisition, technology trend evaluation, IP portfolio creation and strategy, or aligning product development and IP.

To learn more about Global Prior Art and services, see their website www.globalpriorart.com. 

‍

About IPRally

IPRally Technologies is a Finnish software company founded in 2018, with the mission of providing effortless access to technical information, advancing the quality of IP rights and accelerating innovation. IPRally provides explainable AI for patent research and intelligence via an intuitive web platform, enabling intellectual property and R&D professionals to be more productive and make better decisions. The platform is utilized by corporations, patent law firms, academic institutions and public patent offices.

‍

Press Enquiries

Andreas Bask,
PR and Communications Manager, IPRally
andreas.bask@iprally.com
marketing@iprally.com

‍

‍