Understanding IPR and AI
“Innovation transforms a threat into an opportunity, but if its not guarded, the opportunity could turn into vulnerabilities.”
An innovation is understood in terms of intellectual property (IP) which refers to intangible ideas, later shaped into tangible forms such as literary and artistic works, images, designs or symbols. According to the World Intellectual Property Organisation (WIPO), intellectual property refers to creations of mind. The innovators own intellectual property rights (IPR) to protect their innovation from exploitation and vulnerabilities. Intellectual property rights include copyright, trademarks, patents, trade secrets, geographical indications, and industrial designs. Before the digital age, humans exploited IP, but with the emergence of Artificial Intelligence (AI), there are concerns that AI could also misuse intellectual property. Artificial intelligence refers to a computer system capable and smart enough to perform complex human tasks, such as creating, decision-making, and planning. According to NASA, AI systems, perform complex tasks with human oversight, mimicking human behaviour and thought. Machine learning helps them handle cognitive tasks. AI act as an intelligent Robot using perception, reasoning, and decision making to achieve goals. The most widely used AI is the generative AI which refers to AI which generates text, images, speech, video or even technical inventions based on user-inputted instructions; for example, ChatGPT.
Challenges of balancing AI with IPR
Artificial intelligence can be both a friend and foe of humans as it can cater man’s raw ideas and thoughts to innovate intellectual property while, if unethically used then in can lead to the infringement of the already existing intellectual property which is a fruit of absolute human effort. Balancing AI and IPR is a difficult task. If a man uses artificial intelligence to infringe an IPR to create his intellectual property then who is to be prosecuted the man using the AI, the AI itself or the creator of AI? Therefore, with the emergence of AI new avenues for innovation have come into being however, the protection of these innovations is an arduous task due to the emergence of new challenges and complexities. When we talk about AI then ChatGPT is the most widely accepted and used Generative AI (GenAI). ChatGPT came into existence in 2020 and within three months of its launch, it gained over 100 million users making it the fastest growing GenAI. The sole reason behind ChatGPT’s worldwide popularity is it’s free and open access to all. The creators of ChatGPT had a vision of creating a GenAI which can cater to the masses in creation and innovation. However, due to this open access IPR can be easily infringed as there are no rules to regulate the use of AI. However, if AI is overly restricted and protected then it contradicts the creators’ vision of open and easy innovation. Another significant challenge with AI is the unregulated use of copyright data to train AI models that is the text and data mining (TDM) of protected intellectual property leads to the infringement of IPR. Text and data mining can be understood as a breakup of a complex process into three distinct stages: the content is accessed at the first stage, on the second stage, the generative AI extracts or copies the copyrighted content and at the third stage it analyses the data to generate meaningful insights. Thus, the actual TDM happens in the third stage, however, the first two stages raise legal concerns about potential infringement of IPR. Artificial systems use automated text and data mining on copyrighted content to analyse patterns in ideas and information, learning data analysis and formulation to produce valuable and desired outputs. Multiple cases of IPR infringement were observed during AI model training. Various authors, artists and publishers have alleged OpenAI unethically copying copyright data to train its AI model. Similarly, in England, it was claimed that Stability AI infringes copyright by using Getty’s images to train its AI. Other examples include Music publishers accusing Anthropic’ AI unethical copying of copyrighted song lyrics. Programmers have argued that OpenAI misuses codes from GitHub. Moreover, a legal research company claimed that ROSS intelligence copies databases’ case summaries without permission. Intellectual property rights protect human mind creation. However, the IP laws are ambiguous and uncertain about AI’s role in the authorship and invention of the intellectual property. When AI generated content includes protected material like images of real people, then concerns regarding privacy and data protection are raised. Therefore, there are multiple challenges associated with GenerativeAI and the protection of IP. Both the input and output data of the GenerativeAI is prone to exploiting IPR.
How to balance innovation and protection
After looking into the entangled challenges with AI & IPR it is understood that both can be a threat to each other if not property planned and regulated. All these challenges need the immediate attention of the regulating authorities to deal with these challenges. One of the major challenges with GenAI is input data obtained through TDM which infringes IPR. To deal with this challenge the World Intellectual Property Organisation along with the digital experts from across the globe can prepare a digital domain where all the data which is not protected by IPR of whose copyrights or patents have expired can be pooled in together to prepare a humongous database which can be officially and legally used for TDM and AI Model Training. However, the usage of this database should be sanctioned by WIPO with the sanctions of nation specific governments so that the regulating authorities know that which GenAI is using the database for what purpose. The usage of this database officially prepared by WIPO along with digital experts from various nations should be made accessible to AI Training Models with certain rules & regulations, this can further help the regulating authorities to prevent or bring down the rate of cybercrimes abetted by AI & Tech. WIPO with governments of different nations can endow the task of preparing another database to software engineers or computer science professionals, this database should contain a spoof version of the potential copyrighted data and the patented data which can benefit AI Model Training. A cautious approach must be adopted while preparing the spoof database. The software engineers must take due and free content from the IPR owners to make a spoof of their creation. The usage of this database should be limited to GenAI which limit their purpose to education industry. Education is an exception to IPR, so AI education could also be treated as an exception. However, this could be easily missus, so AI education as an exception to IPR should only be introduced with strict limitations, and after careful consideration by the government and regulatory authorities. With the introduction of the task of preparing the spoof database, multiple job opportunities will also emerge for software engineers from across the globe therefore, this as a solution can be looked into. Another challenge with AI and IPR is the ambiguity and uncertainty regarding GenAI’s role in creation of IP. IP laws are silent and do not recognise the input of GenAI and do not give GenAI any credit of authorship of IP created. Thus, the country specific laws should be revisited to clear the air regarding GenAI’s role in creation of IP because, in some cases, the AI output is obtained with a minimal amount of human contribution, therefore, there should be clarity of ownership of IP created with GenAI as well. Henceforth, these are some potential suggestions that we can adopt as solutions to the prevailing entangled challenges of AI and IPR. The recognition of the role of AI as creators of IP with or without human input is extremely significant in the digital age, so that innovation can thrive in today’s competitive digital age.
Conclusion
GenAI and IPR need a systematic balance for creations to survive and flourish in the digital age. AI if judiciously used is the future of innovation while, IPR is extremely imperative to preserve and safeguard these unique innovations. There are multiple challenges entangled with AI & IPR which need to be resolved so that these innovations thrive with safeguard. Therefore, innovations always must be encouraged however, they must be accompanied with some reasonable restrictions which limit the unethical use of innovation of the other innovator.
References
https://www.wipo.int/about-ip/en/artificial_intelligence/policy.html
https://www.nasa.gov/what-is-artificial-intelligence/
https://www.dentons.com/en/insights/articles/2025/january/28/ai-and-intellectual-property-rights
https://academic.oup.com/jiplp/article/19/7/557/7624901
https://www.linkedin.com/pulse/navigating-ai-act-balancing-innovation-intellectual-property-4v81e/