by David Ujhelyi
The World Intellectual Property Organization (WIPO) released its Patent Landscape Report (PLR) on Generative Artificial Intelligence (AI), offering a comprehensive analysis of the patenting activity surrounding this emerging technology. The report delves into patent trends, key innovators, and specific challenges faced by stakeholders in the intellectual property (IP) space as AI continues to revolutionize industries.
Generative AI (GenAI), distinguished by its ability to create new content – ranging from text and images to designs and music – has gained rapid momentum over the past decade, especially after publishing the publicly available models of ChatGPT in 2022, referred to the “iPhone moment” of generative AI development in the report.[1] The patent activity surrounding this technology has surged, with China alone filing over 38,000 patent families,[2] signifying its profound impact across sectors. This rise is propelled by advancements in machine learning, deep learning, natural language processing (NLP), and other AI subfields. Over the past 10 years, the number of patent families in GenAI has grown from just only 733 in 2014 to more than 14,000 in 2023. Since the introduction of the transformer in 2017, the deep neural network architecture behind the Large Language Models that have become synonymous with GenAI, the number of GenAI patents has increased by over 800%. The number of scientific publications has increased even more over the same period, from just 116 in 2014 to more than 34,000 in 2023. Over 25% of all GenAI patents and over 45% of all GenAI scientific papers were published in 2023 alone.[3]
Key Findings of the WIPO Report
1. Patent Filings and Innovation Hubs. The report reveals that the bulk of patent filings in GenAI comes from a few key jurisdictions, including the United States, China, Japan, South Korea, and the European Patent Office. China leads in terms of the volume of filings, while the U.S. excels in high-quality patents. Japan and South Korea also play significant roles, driven by technological developments in robotics, electronics, and telecommunications.[4]
2. Organizations with the most patents. Chinese Tencent, Ping An Insurance Group and Baidu own the most GenAI patents. Tencent plans to add GenAI capabilities to its products such as WeChat to improve the user experience. Ping An focuses on GenAI models for underwriting and risk assessment. Baidu was one of the early players in GenAI and recently unveiled its latest LLM-based AI chatbot, ERNIE 4.0. The Chinese Academy of Sciences (fourth) is the only research organization in the top 10 ranking. Alibaba (sixth) and Bytedance (ninth) are other Chinese companies in the top 10.[5]
3. Publications. The Chinese Academy of Sciences is clearly in the lead in terms of scientific publications with more than 1,100 publications since 2010. Tsinghua University and Stanford University follow in second and third place with more than 600 publications each. Alphabet/Google (fourth) is the only company in the top 20 (556 scientific publications). However, when measuring the impact of scientific publications by the number of citations, companies dominate. Alphabet/Google is the leading institution by a wide margin, and seven other companies are present in the top 20.
4. Technology Sectors Impacted by Generative AI. While the report covers a wide range of industries, certain sectors stand out for their adoption of GenAI. For instance, Software and other applications, Life sciences, Document management and publishing, Business solutions, Industry and manufacturing, Transportation, Security, and Telecommunications.[6]
5. AI Techniques Driving Innovation. The rise in GenAI patents is attributed to advances in machine learning techniques such as deep learning, neural networks, and reinforcement learning. These techniques allow AI systems to generate novel outputs from existing data, such as creating images from textual descriptions or developing unique designs. Patents relating to model architectures, training algorithms, and data preprocessing methods form the backbone of the GenAI IP landscape.
Challenges and Legal Implications
The WIPO report addresses several challenges that arise in the patenting of GenAI technologies. These challenges stem from the complexity of AI systems, the ambiguity around inventorship, deepfakes, and the difficulty of regulating AI-generated inventions.[7]
1. Impact of the labour market. GenAI is expected to significantly impact employment. While some jobs, especially in automation-sensitive sectors, may become obsolete, others will benefit as GenAI enhances productivity, allowing workers to focus on more strategic tasks. It is estimated that 300 million workers in major global economies could be exposed to GenAI-driven automation. Unlike past automation waves that affected middle-skilled workers, this disruption may extend to higher-paid positions. Countries can expect GDP growth, but retraining and support will be essential for affected individuals.
2. Copyright law. Concerns have emerged about copyright infringement involving GenAI art, text, and code, especially regarding AI models using copyrighted material for training without permission. Lawsuits have been filed in the US against AI companies like OpenAI. Another issue is whether AI-generated inventions can be patented. A recent US Federal Circuit decision clarified that inventions created solely by AI are not patentable, but those made by humans with AI assistance are. These legal challenges highlight ongoing IP debates around GenAI.
3. Deepfakes and biases. Concerns about GenAI include deepfakes, which use realistic content to manipulate images or videos, often for malicious purposes like spreading misinformation. Additionally, GenAI models can produce incorrect or biased results due to flawed training data, leading to issues like "AI hallucinations." This makes it crucial for humans to remain involved in decision-making, especially in sensitive industries like finance and healthcare, where accuracy and trust are critical. These risks underscore the need for careful oversight of GenAI applications.
4. Regulation. In response to concerns about GenAI, countries are introducing regulations to balance its benefits and risks. China was an early mover, enacting separate rules for AI types and issuing draft guidelines for industry-wide AI standards by 2026. The EU's AI Act,[8] coming into effect in 2024, categorizes AI by risk and mandates transparency and bias reduction, particularly in high-risk sectors like healthcare. The U.S. has taken initial steps, with an executive order calling for a national AI strategy, focusing on best practices across federal agencies.
The report also provides examples of Responsible AI practices that could help developers making appropriate and ethical decisions, such as transparency and explanations, self-regulating ethical behaviour, avoiding biases, having a human-in-the-loop approach, monitoring ongoing models and building AI models resilient to potential threats.[9]
Conclusion
WIPO’s Patent Landscape Report on Generative Artificial Intelligence offers valuable insights into the evolving IP landscape of one of the most transformative technologies of our time. As GenAI continues to disrupt industries and challenge existing legal frameworks, stakeholders must adapt to new realities in patenting, inventorship, and the ethical use of AI-generated content.
For innovators, investors, and policymakers, the report serves as a roadmap for navigating the complex interplay between technological advancement and intellectual property rights. By addressing the challenges outlined in the report, stakeholders can help ensure that GenAI continues to drive economic growth and societal progress while maintaining respect for ethical standards and legal principles. It is worth noting that the report foresees a great wave of GenAI patents in the future.[10]
[1] Report p. 14.
[2] Report p. 9.
[3] Report p. 7.
[4] Report p. 43.
[5] Report p. 52.
[6] Report p. 57.
[7] Report p. 67–69.
[8] Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending Regulations (EC) No 300/2008, (EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828 (Artificial Intelligence Act)
[9] Report p. 69.
[10] Report p. 70.
