Home Insights Generative AI and its impact on the IP system

Generative AI and its impact on the IP system

Read time
4  minute read
Date published
13 July 2023

Griffith Hack managing director Aaron LePoidevin takes a look into the challenges, opportunities and potential overall impact of generative AI technology on the IP system.

As an avid follower and participant in the tech and startup ecosystem, I’ve been fascinated by the ongoing conversations around generative AI and its potential impact on the global IP system.

According to a recent McKinsey report, generative AI’s impact on productivity could add trillions of dollars in value to the global economy annually. It seems that in the blink of an eye, generative AI has moved from hypothetical disruptor to unavoidable reality.

And while AI’s impact on how we do business is starting to be quantified and mapped out, its impact on the innovation and creative process remains less clear – and that is why it’s such an intriguing and important topic in the context of IP.

AI, innovation and intangible assets

AI is being employed across a wide range of industries, impacting almost every aspect of the innovation and creative process. The availability of large amounts of training data and advances in affordable high computing power are accelerating Al’s growth, allowing it to achieve breakthroughs that seemed unlikely only a few short years ago. With the ability to create novel and realistic content such as images, music, and text, and the very real possibility of performing technical innovation, generative AI has the potential to revolutionize multiple industries. As a result, AI raises diverse IP questions.

IP is an important economic asset. Data and intangible assets are the foundation of our digitalized world and as a result, IP is rapidly gaining importance over physical assets. The number of IP filings has been increasing in recent years but is also subject to more volatile swings. Markets are globalizing and intangible assets are flowing much more freely across borders. In an uncertain and diverse world, an accessible, balanced, and effective IP system should help people prosper from good ideas.

AI and the global IP system

In order to harness the opportunities of generative AI, we need a global IP system that continues to support and foster innovation and creation, and for the underlying systems of IP administration to evolve in-line with the swift adoption of new technology like generative AI.

The IP system that we have today was developed when the speed of innovation was much slower and and focused on human creation and innovation. It encouraged the development of technologies that largely enabled the production process and trade of physical products.

Collectively we need to ensure new technologies make IP and the IP system easier to access, more efficient and more understandable, noting the risks that they also pose.

What are the current challenges?

AI inventions present the current patent system with a number of challenges. Let’s explore a couple of examples:

AI-assisted inventions and the patent process

In an ever-growing number of jurisdictions, it is accepted that technologies created solely by AI systems – without any human input – cannot be patented. We recently saw significant legal decisions handed down in both Australia and New Zealand that determined that inventions must list a human in the patent application.

However, this does not entirely address the legal uncertainties that exist in situations where inventions are jointly created by both humans and AI systems. With the rapid advancement of generative AI, such collaborations are likely to become more common. This is particularly hazardous in the United States for example, where patents can be challenged for failing to list all inventors. Without a mechanism to acknowledge the contributions of AI systems, it opens the door to an increased number of challenges.

My colleague Anthony Selleck published a fascinating article recently that shed light on the uncertainties and challenges of generative AI’s impact on patent systems – but it also suggested a path forward. I agree with Anthony’s proposition that international cooperation between patent offices and legislatures is required to establish a common set of guidelines for AI-assisted inventions. As Anthony states in his article: “It is only through international cooperation and legal harmonisation that inventors and patent practitioners can have the certainty they need to step into this new frontier in innovation.” It will be fascinating to see how patent offices around the globe proceed in the coming months and years – we’ll be watching very closely.

Streamlining the patent process

Generative AI may have the ability to reduce friction in the patent process itself. From assisting in preparing patent specifications to comparing inventions with prior art and meeting other legal requirements, AI could augment these steps – even if the initial invention was developed by a human. This augmentation could enhance accessibility to the patent process, potentially driving up its usage and democratising innovation. It also poses the risk of flooding the IP system with an exponentially increasing number of AI-generated patent applications, placing strain on the system.

And AI’s involvement in the patent process doesn’t end with the initial application. It could aid in the development of alternative solutions to patented inventions that avoid infringement. AI could also assist in drafting patents by better identifying and covering the key concepts, thereby improving patent quality and making them more robust. Additionally, it can contribute to automating infringement searches, helping to identify potential infringement risks efficiently.

Commercialisation strategies:

AI’s capabilities extend to the commercialisation and monetisation of patents as well. Automated matching of technologies, identifying infringement risks, validating patent strengths, and uncovering market trends are just a few ways in which AI can bolster IP strategy and enhance the value of patent portfolios. By harnessing the power of AI, organisations can make informed decisions regarding licensing, partnerships, and enforcement of their IP rights.

Ultimately, the challenges presented by generative AI must be addressed to strike a balance between promoting innovation and preserving the integrity of the IP system.

It was pleasing to see IP Australia undertake a proactive discovery process to better understand the potential impact of generative AI on the IP system. You can read about their findings, published earlier this month, here.

If you have experimented with generative AI, please send me an email. I’d love to hear about your experiences.