AI data sharing, copyright...
What do you think? Could we have a registration system like patents, with an eye toward revision of the Berne Convention?
This is more in the sense of data aggregation to AI, from the perspective that it would be better to aggregate data sets.
Of course, copyrights can be obtained as is without any method, but it is impossible to exercise rights when the data is imported into AI (at least in the case of Japan, there are restrictions on rights).
If you set up an organization or something (an international organization would be better) to create a data set and provide your work to it, you will receive some compensation.
The organization will provide the dataset at a relatively low cost and basically indiscriminately.
Of course, A governance would need to be in place here.
If the dataset containing the work in question is good and sells well, the organization will provide additional compensation to the work provider, thereby promoting the provision of good works.
International organizations without an association, of course, should be non-profit organizations that can provide datasets at a reasonable price (more expensive than open－source, but with the advantage of being free of errors) and that can also provide some educational services.
In the area of intellectual property, the law of patents allows for international applications to be filed in batches (only the application is filed, and the examination is done in each country. (Only the application is filed, and the examination is conducted in each country. The principle of patent independence applies).
I wonder if it would be possible to make such a system include AI governance as well.
Just as in the EU, there is also a "right of pursuit". This is one of the non-transferable moral rights, which is not recognized in Japan, and in simple terms, when the price of an artwork increases significantly from the initial price, the right holder gets part of the compensation from the amount of the significant price inc－.
I have a feeling that AI data sets can also be used in a certain way.
If this can be put together well, I think it would be a good institutional proposal...
Could someone please take this up?
