Going by the article you linked, game cloning runs into problems, including the EULA and potentially for copying information. Copyright cases also consider whether the potentially infringing text replaces the original in the market, which is another problem for cloning games.

Training models, as a concept, isn’t necessarily IP theft, but in practice has involved quite a bit of IP theft. Anthropic itself has come under fire for pirating books and scraping Reddit after agreeing not to scrape Reddit.

Ah, it’s ok to steal IP to build models, but not to use those models to steal IP. Got it.

You’re delving into the unpleasant realm of risk assessment. How likely is the app going to be targeted by a threat actor, and how badly could they use it to damage your data or system?

If it’s a public facing service, I’d immediately check if I have alternative applications I could migrate too. Any hold an outsider gets through the public facing app could be used to pivot further into my network.

If it’s something that doesn’t interact with the network at all, then I wouldn’t worry nearly as much, but might look for alternatives any just because I know this isn’t going to get improved at all. Somewhere in the middle, you need to decide how much risk you can accept.