This sounds like the exact opposite of what the title is claiming.
- As there are no external motives (eg. for profit) held by the AI when analyzing the works it is not a copyright infringement.
- Parsing through whether the data set is legally obtained is not feasible therefore that shouldn't be the limiting factor to further development.
However as the transcript was uploaded last month, things may have changed since then.
Japan was behind about developing search engine service. Some people argued that it's due to copyright law (there's no fair use) and it shouldn't be repeated again. So the gov want to encourage AI developing by explicit law.