January 29, 2021
Terminology such as Artificial Intelligence (AI) and Machine Learning (ML) have become commonplace – much like “text message,” “email,” and “smart device.” Turn on the news and you are very likely to hear such terms in various capacities. Such new-found popularity is bound to cause industries to revamp their thinking; their organization; their goals. It is also bound to instill unease in those concerned that the machines will take over.
As with any rapidly growing technological area on the industry side, there is also a rapidly growing number of patent applications being filed on the intellectual property side. As a result, the five largest intellectual property (IP) offices in the world – the European Patent Office (EPO), the Japan Patent Office (JPO), the Korean Intellectual Property Office (KIPO), the National Intellectual Property Administration of the People’s Republic of China (CNIPA), and the United States Patent and Trademark Office (USPTO) – have collectively set up a special task force to coordinate their initiatives in the AI space. According to the EPO’s website, the task force is “exploring legal, technical and policy aspects of new technologies and AI, their impact on the patent system and on operations at the five offices.”
Each of the aforementioned offices has also taken individual initiatives to address this ever-growing market. For example, the EPO held its first public conference on AI in May 2018. Later that year in November, the EPO issued guidance for examination for AI and ML patent applications. To briefly summarize, the EPO’s guidance indicates that subject matter that “is directed to a purely abstract mathematical method and … does not require any technical means” is not suitable for patent protection. However, “[i]f a claim is directed either to a method involving the use of technical means (e.g. a computer) or to a device, its subject-matter has a technical character as a whole and is thus” eligible for patent protection.
Further, the EPO’s guidance explains that AI and ML are based on computational models, which “are per se of an abstract mathematical nature, irrespective of whether they can be ‘trained’ based on training data.” Therefore, the EPO instructs its examiners to “carefully” look at certain expressions within a claim (i.e., trigger words) -- such as “support vector machine,” “reasoning engine,” or “neural network” because they “usually refer to abstract models devoid of technical character.”
According to the EPO’s guidance, the key for patent eligible subject matter is whether such subject matter is tied to something “technical” or has a “technical purpose.” As an example, the EPO explains that “the use of a neural network in a heart-monitoring apparatus for the purpose of identifying irregular heartbeats” and the classification of “digital images, videos, audio or speech signals based on low-level features (e.g. edges or pixel attributes for images)” have a technical purpose or application, and would thus be eligible for patent protection. By contrast, “[c]lassifying text documents solely in respect of their textual content” does not serve a technical purpose, but rather a linguistic one. The EPO also indicates that “[c]lassifying abstract data records or even ‘telecommunication network data records’ without any indication of a technical use being made of the resulting classification is also not per se a technical purpose.” As such, these types of classifications would not be eligible for patent protection.
Since the aforementioned guidance back in November 2018, the EPO has continued its efforts in this field by creating a dedicated team in 2019 with the goal of applying AI and ML “to increase efficiency and quality in the patent grant process.” The team has six data scientists as core members and is supported by patent examiners – the focus being primarily on three core AI projects – Natural Language Processing, Computer Vision, and Machine Translation. This further shows the EPO’s dedication to AI not only for determining subject matter eligibility for patent protection in this space but also for improving its own internal processes such as classification, search, and machine translation.
For those interested in the technical details, the EPO’s website provides the following description of the AI it uses:
EPO AI uses state of the art deep learning network architectures and adapts them to handle the challenges of the patent domain. The EPO’s core language models are trained on millions of documents stored in the EPO’s prior art databases and are fine-tuned for addressing the complexities of the patent domain, such as technical language and syntax. AI at the EPO today is driven by supervised machine learning using the previous work of our highly skilled examiners.
Fast forward to 2020, even during the pandemic, the EPO continued its interest in AI – although virtually. In September, the EPO held a Tech Day event in which participants discussed the opportunities and challenges of AI from both the technological and law perspectives. In December, the EPO held a two-day conference in which policymakers, investors, inventors, small and mid-size enterprises, academics, and IP professionals exchanged views and expertise on AI and IP rights. Notable participants included members of the EPO, representatives of the JPO, CNIPA, KIPO, and USPTO, as well as various industry insiders.
During the December conference, the EPO also detailed the latest initiatives relating to AI and presented its most recent study entitled “Patents and the Fourth Industrial Revolution;” the Fourth Industrial Revolution (4IR) being a global trend that is “driven by a constellation of disruption technologies which together are paving the way to a data-driven economy.”
The study, which may be found here: https://www.epo.org/service-support/publications.html?pubid=222#tab3, is intended to serve as a guide “through a major technology transformation that impacts a wide range of sectors of the economy.” The key take-away points of this study were as follows:
(1) 4IR innovation has dramatically accelerated during the past decade and accounted for more than 10% of global innovation in 2018.
(2) The US remains the world leader in 4IR technology, despite the fast growth of 4IR innovation in Korea and China. Europe is losing ground to other global 4IR innovation centers, despite the remarkable performance of small countries such as Sweden and Switzerland.
(3) The dynamism of national industry champions and regional clusters in 4IR technologies explains the domination of the US and the rise of Korea and China in the 4IR innovation landscape. By contrast, the relative weight of the top European and Japanese 4IR applicants has diminished since 2010, while the main 4IR clusters in Europe and Japan have experienced slower growth in their innovative activities.
It is undeniable that AI has moved to the forefront of our daily lives and it is reassuring that the EPO, along with the other largest IP offices, have taken the necessary steps in understanding and fully embracing this technology. While new emerging technologies are always exciting (I equate this to the thrill of first unwrapping a brand new smart device), there are many people who fear that AI may take over our lives – whether it is our jobs, our homes, our vehicles, our thoughts.
Similar to the other large IP offices, the EPO has at least attempted to address this concern – best summed up by Nellie Simon, Vice President of the EPO’s Directorate-General Corporate Services – “The future is about humans standing on the shoulders of machines, not being replaced by them.”