AI systems rely heavily on data. The access to data and the use of data to set up AI systems can cause intellectual property problems when the data is protected by copyright and related rights. Accordingly, artificial Intelligence and Machine Learning are not just for Big Business but it’s a problem that all businesses will embrace in the coming future.
In computer science, artificial intelligence or AI is intelligence demonstrated by machines, as opposed to the natural intelligence of humans. Typically, the term “artificial intelligence” is often used to describe machines (or computers) that are capable of mimicking the “cognitive” functions that humans would normally associate with the mind, like “learning” and “problem-solving”.
In the 21st century, AI techniques have experienced a resurgence following simultaneous advances in computer power, big data, and theoretical understanding. Subsequently, AI engineering has become an essential part of the technology industry, helping to solve many challenging problems in machine learning, software engineering, and operations research.
Artificial intelligence (AI) has been applied in many different fields such as labor, health care, security, traffic, marketing, etc. However, one of the common misconceptions is that AI technology and machine learning (ML) technologies are considered suitable only for large corporations and not feasible for small businesses or startups because of security risks and high operation costs. With the advancement of technology in the industrial era 4.0 and consequently, the gradually declining cost of this technology, AI and ML for small and medium businesses are now slowly becoming a reality. Whether in the form of virtual assistants, chatbots, or automated robots, AI is giving businesses an immense advantage by improving productivity and efficiency, sales quality, and even human resources, etc.
Specifically, AI helps extract sales-related insights from customer data, allowing business owners to analyze customer feedback and tailor sales activities, reaching the right target customer. It also improves customer care and communication through the use of automated chatbots or virtual assistants, modernizes inventory operations and processes, and delivers high-reference-value projections, etc.
Some free or low-cost smart applications such as Bizfly Chat, VietGuys Ai Call Center, Amis.vn, etc. are regularly becoming more popular. Small and medium business leaders can consider applying these applications for their businesses to have a more productive working system. Additionally, Machine Learning tools like Google TensorFlow and Spark Mllib can help businesses by providing valuable insights into customer behavior based on data online interactions, online purchase history, online transaction type. Furthermore, these applications can also provide relevant product recommendations to small e-commerce stores based on ML tools that learn from previous customer purchases and selections, etc.
Role of Intellectual Property regarding AI
For SMEs, it is important that they make AI and ML a core element to their operating system and apply the right intellectual property (IP) strategy to protect them.
AI systems rely heavily on data and accordingly, during the process of setting up the AI, AI engineering needs to access and use the data which is probably consisted of images, songs, or other people’s works. However, nearly all of these data are protected by copyright and related rights. In addition, structuring an artificial neural network, mimicking a patented process, could also be considered patent infringement.
Enterprises may also face other problems such as the protection and enforcement of AI-generated intellectual properties. In addition, if needed, enterprises also need to support the creation and improvement of existing legal systems related to the provisions of AI because currently, the law has not kept pace with the advancement of technology.
To deal with this problem, businesses, especially small and medium-sized businesses that apply AI and ML in their production and business activities, must have an effective IP strategy to maximize profit from investment and research for AI and at the same time, gain a competitive advantage over competitors.
The ways to protect AI and ML of businesses
The methods for strengthening the harmonization between AI and IP may consist of setting up agreements, contracts of assignment for data containing the IPR of others. In addition, data sets created by enterprises themselves for AI or ML practice need to be identified, labeled, and apply security measures and restrict access.
Businesses can also apply similar measures to principles, structures, algorithms, methods that can help ML to develop self-learning capability which can make AI and ML aggregate data, analysis, and predictions. However, this method is not optimal because it limits the cooperation and development of AI and ML.
In addition, if an enterprise creates its software using AI, it is advisable to register for copyright protection of that software right away because protecting the AI system is sometimes more important than the products created by the AI system themselves.
Besides, participating in open source projects in the short term helps to adopt and develop AI technology faster, save costs but without the ownership of that projects can have a negative impact in the long term. Therefore, businesses need to consider developing their own source code while remaining involved in the open-source community by choosing the right license terms.
Enterprises also need to make a list of their intellectual property assets in general and AI-generated assets in particular in order to manage, protect and exploit effectively and in accordance with current regulations. For example, technical solutions, new product designs, design products, even command lines that execute AI-generated software that cannot yet be registered for IP protection may be protected with security measures such as with the protection of business secrets. At the same time, build strong and detailed transaction contracts related to the input data and output products of AI to fill the gaps caused by the lack of a legal framework to protect these objects.
The importance of improving the legal system of intellectual property related to AI
Like most developed countries in the world, Vietnam is investing more and more and the development of Artificial Intelligent – the key point to the future in this technology era. Subsequently, Vietnam aims to become a center of innovation, development of AI solutions and applications in the ASEAN region as well as in the world.
For the policy and legal system, Vietnam has an IP Strategy to 2030 and a National Strategy on Research, Development, and Application of AI to 2030. This shows the important role of AI and IP in the development of the country. To realize this strategy, it is necessary to have a simultaneous implementation of solutions, such as completing legal documents on intellectual property rights related to AI, promoting research, training, communication, and building a comprehensive AI database infrastructure, etc. In which, it can be seen that the completion of legal documents on IPR related to AI is brought to the forefront. Once the legal framework has been completed, businesses can protect and exploit AI-related intellectual property effectively and transparently.
And to answer the question that many skeptical entities still raise about the necessity of harmonizing AI and IP and the need to complete legal documents on intellectual property rights related to AI, there is one example, a minor conflict between the EPO and Dabus in 2019 that will clearly demonstrate that the world is, in fact, in an urgent need to deal with the matters raised between AI and IP.
In 2019, the European Patent Office has refused to protect a patent for a food container created by an artificial intelligence entity named Dabus on the grounds that Article 58 of the European Patent Convention stipulates that the inventor must be human. This caused a wave of debate on the question that should I protect the intellectual property of AI’s creative output?
This is a vivid example of the constant movement of the world of technology that has challenged the stability of the legal system. Therefore, the IP system needs to react flexibly and sensitively, minimizing negative impacts while ensuring the encouragement of innovation and new technologies.