The famous scientist Alan Turing (also known for his important contribution to breaking the Enigma Code used in World War II) defined a simple test (known as the “Turing test”) for determining whether Artificial Intelligence (AI) can “think”. This theory says: if a human observer has a conversation with an AI device and cannot differentiate the AI from the human or even believes the AI is human, the AI can think like a human being.

 

For more information on the Turing test see here

 

Will human intelligence be replaced with AI?

 

This question is at the centre of much debate, with contributions from non-fiction authors like Yuval Noah Harari, fiction novelists like Ian McEwan and Frank Schätzing, or scientists like Hiroshi Ishiguro who develops androids.

 

The present definition of what is or is not human may lag behind the development of AI and its future significance. However it is not unreasonable to expect that science fiction will inform what becomes reality.

 

In law, human rights still prevail

 

The European Patent Office recently refused two patent applications, EP 3 564 144 and EP 3 563 896 on the grounds that an AI computer device called DABUS was designated as inventor, “a type of connectionist artificial intelligence”.The applicant stated that he had acquired the right to the European patent as the employer of DABUS and thus as a successor in title. The applicant explained that the machine “identified the novelty of its own idea before a natural person did” (for more information on these decisions, see here and here).

 

For further discussion of these decisions, please see here: IP Rights in the Context of Artificial Intelligence 

 

During oral proceedings in a consolidated procedure, the applicant argued that the legislators did not exclude patenting of inventions made by an AI system. Furthermore, he described DABUS as the actual devisor of the invention underlying the applications and thus as truthful inventor. As the owner of DABUS, the applicant claimed he was to be considered as successor of rights.

 

The Examining Division (ED) concluded that the designation of a machine as inventor did not meet the formal requirements under Art. 91 and Rule 19(1) EPC.

 

The ED stated that the EPC, in the context of inventorship, only made reference to natural persons, enabling them to be identified by name and thus to exercise their rights. The name therefore formed a key part of their personality. “Things have no rights which a name would allow them to exercise.” And “AI systems or machines have at present no rights because they have no legal personality comparable to natural or legal persons.”

 

In addition, the ED stated that “AI systems or machines can be neither employed nor can they transfer any rights to a successor in title”.

 

As a result, the applications were refused, with the possibility of an Appeal.

 

We will be interested to learn when AI will be allocated rights – or autonomously claim its rights, or simply act like a human as in McEwan‘s “Machines like me” …

 

 

For more information, please see our page onArtificial Intelligence