3 ms·
The most fitting method would be to be to train an LLM on the Claude Code source-code (among other data). Then use Anthropic's own argument that LLM output is
by aeternum 7mo ago
The most fitting method would be to be to train an LLM on the Claude Code source-code (among other data).
Then use Anthropic's own argument that LLM output is original work and thus not subject to copyright.