At the 5th meeting of the Subcommittee on Legal System of the Copyright Subcommittee of the Council for Cultural Affairs held on December 20, 2023, the Agency for Cultural Affairs published a “Draft Approach to AI and Copyright”.

This article introduces some of the Agency's current thinking on the interpretation of Article 30-4 of the Copyright Act and its scope of application.

Article 30-4 of the Copyright Act

(Exploitation without the Purpose of Enjoying the Thoughts or Sentiments Expressed in a Work)

Section 30-4

It is permissible to exploit a work, in any way and to the extent considered necessary, in any of the following cases, or in any other case in which it is not a person's purpose to personally enjoy or cause another person to enjoy the thoughts or sentiments expressed in that work; provided, however, that this does not apply if the action would unreasonably prejudice the interests of the copyright owner in light of the nature or purpose of the work or the circumstances of its exploitation:

(i) if it is done for use in testing to develop or put into practical use technology that is connected with the recording of sounds or visuals of a work or other such exploitation;

(ii) if it is done for use in data analysis (meaning the extraction, comparison, classification, or other statistical analysis of the constituent language, sounds, images, or other elemental data from a large number of works or a large volume of other such data; the same applies in Article 47-5, paragraph (1), item (ii));

(iii) if it is exploited in the course of computer data processing or otherwise exploited in a way that does not involve what is expressed in the work being perceived by the human senses (for works of computer programming, such exploitation excludes the execution of the work on a computer), beyond as set forth in the preceding two items

Article 30-4 of the Copyright Act stipulates that “when the purpose is not to personally enjoy or cause another person to thoughts enjoy or sentiments” (hereinafter referred to as “non-enjoyment purpose”), the use may be made without the permission of the copyright holder (Article 30-4, Pillar 1 of the Copyright Act). Paragraph 2 of the same Article states “use in data analysis”. Therefore, under the current Copyright Law, it has been understood that any use of material for data analysis, including the training of generative AI, falls under the category of a non-enjoyment purpose and thus can be carried out without the permission of the copyright holder.

However, there can be cases where multiple purposes coexist in a single act of use. In such a case, even if one or more purposes of a certain act of use falls under the category of “use in data analysis” or another non-enjoyment purpose, if any of the purposes is deemed to be an enjoyment purpose, the requirements of Article 30-4 of the Act are not met and the rights limitation provision of Article 30-4 of the Act may not be applied.

In the case of generative AI, the following cases can be assumed as specific examples of cases in which the an enjoyment purpose is regarded as coexisting with a non-enjoyment purpose.

(1) Reproduction, etc. of a work for the purpose of fine tuning with the aim of outputting the learning data as it is

(2) Reproduction of works for the purpose of fine tuning to output products strongly influenced by learning data

(3) Reproduction, etc. of a work for the purpose of creating a database in which the contents of a work are converted into vectors for the purpose of outputting all or part of an existing database or data posted on the Web using a generative AI , even though there is no intention to output training data used for AI learning.

Article 30-4 of the Copyright Act also covers the use of works for learning in order to develop AI other than generative AI (AI that performs recognition, identification, human judgment support, etc.), use of works for technological development and practical use testing, reverse engineering of programs, and other acts.

Summary

Article 30-4 of the Copyright Act was established as a right limitation provision to flexibly respond to new modes of use of copyrighted works in line with technological innovation. At the time of the legislation, it was believed that AI learning was intended for Internet search services and the like. However, the recent technological progress of generative AI has been tremendous, and it is becoming widely used not only by businesses but also by the general public, which has led to an increasing number of problematic situations regarding the interpretation and scope of application of the Article. Therefore, there is a need to formulate rules and reorganize the scope of application in anticipation of generated AI.

According to a draft proposal recently released by the Agency for Cultural Affairs, the direction is that even in the case of having a generative AI learn a copyrighted work, permission of the right holder is required when, for example, the purpose is to output a part of the copyrighted work.

According to the Agency's schedule, public comments will be made during January 2024, and a report will be made at the Copyright Subcommittee of the Council for Cultural Affairs in March of the same year. We plan to provide further updates in our newsletter.