At the end of July, the patent world was shaken by two news: patents officially attributed to artificial intelligence (AI), called Dabus, were issued in South Africa and Australia. This comes after the European Patent Office (EPO) back in 2019 that AI could not be the author of an invention because it required legal personality. The South African decision is not revolutionary – the Patent Office of this state only deals with formal requirements of documents, so theoretically such a patent with AI as the author could be issued in Lithuania too. Meanwhile, in Australia, a court has already dealt with the issue and said that Australian law does not define the inventor of an invention only as a human being, and that the possibility of a patent for an invention developed by AI is in line with the current reality of technology and promotes innovation.
The UK and US authorities came to a different conclusion – the invention was not patented in those countries.
This means that a company that uses AI to create its inventions will have the same ability to patent its invention as a company that uses only human intelligence. This is a big step forward in innovation, but isn’t it going to be that AI, whose learning skills are limitless and don’t need lunch or weekends, will gain momentum and push the real inventors aside?
Another important moment is how the state of the art have to be assessed in the future?
It is clear that AI’s skills will exceed the knowledge of a specialist in the relevant field, so that the assessment may no longer fulfil its primary purpose in patent law.
This situation is particularly important in the pharmaceutical industry, where AI is increasingly being used to invent effective, cheaper and faster medicines.