Hi pinoaffe, Thanks for your feedback!
Regarding potential legal and ethical issues you mention I think everyone will have to make their own judgment before using this technology. The company I am working for has recently decided to deploy GitHub Copilot for a large population of developers (multiple thousand licenses) and they deemed it to be OK and I also did.
Regarding the technical issue you mention:
- LLMs are very good at generating text/code that *looks* plausible/correct, but that is not necessarily correct. Therefore, it is often more difficult to verify that code generated by an LLM is not subtly broken than it is to write similar code yourself. *PLEASE* do not underestimate this issue.
Of course the results must always be evaluated by the developer to ensure correctness. In my experience, there are sometimes cases were the results generated by LLMs are not correct. I have seen this both with ChatGPT as well as GitHub Copilot. However, in my experience the positive results that I got definitely outweigh the negative results.
Best regards, Tobias _______________________________________________ sword-devel mailing list: sword-devel@crosswire.org http://crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
