What is ChatGPT?
The OpenAI organization has created ChatGPT, i.e. an advanced chatbot using artificial intelligence, based on the Generative Pre-trained Transformer 3 (GPT-3) machine learning model.
ChatGPT can generate a natural-sounding answer to a user's question in real time. ChatGPT works in multiple languages (including Polish) and uses 'knowledge' extracted from Internet resources. It is capable of answering user-submitted questions in an accessible manner and of holding a conversation with the user on any topic, and is also capable of generating text according to user-specified instructions.
ChatGPT was launched on 30 November 2022 and has been a global success, with more than one million users within the first five days of its launch.
Example use of ChatGPT
First of all, ChatGPT can be used to generate answers to questions, create articles, reports or social media posts, as well as other textual content within customer service systems. ChatGPT can also be used to help write a computer programme, as it is familiar with programming languages.
According to the Fishbowl survey (available at link: fishbowlapp.com), marketing professionals (37% of respondents) and IT professionals (35% of respondents) are the most likely to use ChatGPT. This was followed by professionals from the financial sector and lawyers.
ChatGPT and intellectual property law
The emergence of such advanced tools based on artificial intelligence poses challenges that require appropriate regulation to ensure the protection of intellectual property rights and to set a framework for the use of such tools in specific sectors, e.g. education.
Doubts should also be raised about the possibility of granting protection to content (text, graphics or video material) created using artificial intelligence-based systems.
There are different approaches to this problem around the world, but the prevailing view today is that only man-made works can be protected under copyright law and, given the high creative autonomy of artificial intelligence systems, there is no human connection between the generated work and humans. The inability to grant copyright protection to content generated by artificial intelligence systems, according to some, points to the need for separate legislation under which such works would be protected.
Another challenge for regulation is the issue of infringement of intellectual property rights by artificial intelligence systems. This is because there is a risk that these systems may (unintentionally or intentionally) generate content that infringes the intellectual property rights of others.