Also, don't use AI to translate into a language you can't read back. It glosses over assumptions you didn't know you made, and the output might not mean what you intended.
We used to have machine translation that worked well enough for readers to know the overall meaning most of the time, but when it failed, it failed spectacularly and the reader would know not to trust the output.
Now with LLM translations we can’t trust anything. Even simple translations which used to work perfectly now shoves extra LLM-isms into the translation making the prose ugly.
sasaf5 · · focus · HN ↗
runarberg · · focus · HN ↗
We used to have machine translation that worked well enough for readers to know the overall meaning most of the time, but when it failed, it failed spectacularly and the reader would know not to trust the output.
Now with LLM translations we can’t trust anything. Even simple translations which used to work perfectly now shoves extra LLM-isms into the translation making the prose ugly.