AI Times, a Korean AI publication, has published Managing Partner Yongduck Kim's column on AI patents.
Yongduck Kim, a patent attorney, focuses on patents in cutting-edge technology fields such as artificial intelligence (AI), smart factories, and blockchain, and has experience performing intellectual property (IP) work for various technology companies such as Samsung Electronics, LG Electronics, and Xiaomi. In addition, he has written books such as “AI Patent Examination Practice Guide” and “AI Patent Strategy” and continues to share his expert knowledge on AI technology and patent strategy.
In this column, we are dealing with the issue of ‘AI conversation transparency’, which has emerged in the generative AI era, under the theme of “AI writing, human speaking... SAP patent’s AI communication differentiation technology”. LLM is creating a natural dialogue that is difficult to distinguish from human beings, and the scope of use in various fields such as customer consultation, education, finance, medical care is rapidly expanding. But the challenge of trust, accountability and consumer protection is emerging as users find it difficult to understand whether they're talking to real people or with artificial intelligence.
The column analyzes the technical structure of the SAP SE's U.S. registered patent US 12,438,836 B1 (Detection of whether a communication is generated via artificial intelligence) to determine whether an online chat partner is human or AI. The patent goes beyond simply analyzing the conversation content, and proposes a multiple test cluster structure that identifies the communication structure of the chat interface and analyzes the response speed, language characteristics, context response, error-induced response, etc., after the test application forms a direct communication channel with the other party. The key is to be able to mechanically detect the unique response patterns of the generative AI.
In particular, this column notes that not only “technologies that make AI good” but also “technologies that determine AI” are coming to an important era. As generative AI becomes more human-like, so does the importance of technologies that allow users to verify and transparently verify AI interventions. It also sheds light on how these technologies can be linked to future AI transparency regulations, consumer notice obligations, and accountability for automated decision-making, and what they mean from a patent perspective.
The full article can be found at the link below.
This article reflects the information available when it was published. Contact us to discuss your circumstances.