Between 2014 and 2023, China filed an impressive 38,000 generative AI-related patents, significantly outpacing the USA’s 6,276 patents, out of a global total of 50,000 patents, according to the World Intellectual Property Organisation (WIPO).
Notably, 25% of these 50,000 patents were filed in the past year alone.
“This is a booming area that is growing at increasing speed. We expect it to grow even more,” says Christopher Harrison, WIPO Patent Analytics Manager. “The patent data suggests this is an area that will have a profound impact across many different industrial sectors”.
Key Insights and Trends
The third largest filing country is Korea, followed by Japan and India. Notably, India is filing patents at a faster rate than any other country, according to WIPO.
From 2014 to 2023, there were 54,000 generative AI-related inventions (patent families) filed and over 75,000 scientific publications published. The growth in patents has been rapid, with an eightfold increase since 2017, coinciding with the introduction of deep neural network architecture behind the Large Language Models now synonymous with generative AI source.
Suggested Diagram 1: Global Patent Filing Trends (2014-2023)
- Place this line graph here to illustrate the annual number of generative AI patents filed globally, highlighting key technological developments.
Yearly Growth and Patent Distribution
In 2023 alone, over 25% of all generative AI patents globally were published, along with over 45% of all related scientific papers. Despite this growth, generative AI patents currently represent only 6% of all AI patents worldwide.
Suggested Diagram 2: Top 10 Organizations by Patent Count
- Insert this bar chart here to display the number of generative AI patents filed by the leading organizations.
Leading Organizations and Inventor Locations
The top 10 organizations filing generative AI patents are:
- Tencent: 2,074 inventions
- Ping An Insurance: 1,564 inventions
- Baidu: 1,234 inventions
- Chinese Academy of Sciences: 607 inventions
- IBM: 601 inventions
- Alibaba Group: 571 inventions
- Samsung Electronics: 468 inventions
- Alphabet: 443 inventions
- ByteDance: 418 inventions
- Microsoft: 377 inventions
Top five inventor locations:
- China: 38,210 inventions
- USA: 6,276 inventions
- Republic of Korea: 4,155 inventions
- Japan: 3,409 inventions
- India: 1,350 inventions
Types of Data and Sector Distribution
Generative AI patents predominantly involve image and video data (17,996 inventions), followed by text (13,494 inventions) and speech/music (13,480 inventions). Patents using molecule, gene, and protein-based data are growing rapidly, with 1,494 inventions since 2014 and an average annual growth rate of 78% over the past five years.
Suggested Diagram 3: Patent Distribution by Data Type
- Include this pie chart here to illustrate the proportion of generative AI patents by the type of data they utilize (image/video, text, speech/music, molecule/gene/protein).
These patents span across various sectors, including:
- Life sciences: 5,346 inventions
- Document management and publishing: 4,976 inventions
- Business solutions, industry and manufacturing, transportation, security, and telecommunications: Over 2,000 inventions in each sector
Suggested Diagram 4: Sector Distribution of Generative AI Patents
- Add this bar graph here to show the number of patents filed across different industrial sectors.
Future Prospects
Generative AI holds immense potential for the future. It can aid in designing new molecules, expediting drug development, automate tasks in document management and publishing, enhance retail assistance systems and customer service chatbots, and enable new product design and optimization, including in public transportation systems and autonomous driving, says WIPO.
This comprehensive overview highlights China’s leading position in generative AI patents, the rapid growth in this field, and its potential impact across various sectors.
