Forbid LLM contributions (#3703)

This commit is contained in:
Natalie Weizenbaum 2023-09-27 15:15:00 -07:00 committed by GitHub
parent 8192ca4c1c
commit ce83bcebbc
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -16,6 +16,7 @@ process][], very small features can follow the [fast-track process][] instead.
* [Embedded Protocol](#embedded-protocol) * [Embedded Protocol](#embedded-protocol)
* [Fast Track](#fast-track) * [Fast Track](#fast-track)
* [Emergency Track](#emergency-track) * [Emergency Track](#emergency-track)
* [Large Language Models](#large-language-models)
## Process ## Process
@ -371,3 +372,14 @@ as follows:
2. These pull requests may be merged as soon as they're approved. If the issue 2. These pull requests may be merged as soon as they're approved. If the issue
appears outside of work hours, it may be merged without review, but a *post appears outside of work hours, it may be merged without review, but a *post
facto* review should be done as soon as possible. facto* review should be done as soon as possible.
## Large Language Models
Do not submit any code or prose written or modified by large language models or
"artificial intelligence" such as GitHub Copilot or ChatGPT to this project.
These tools produce code that looks plausible, which means that not only is it
likely to contain bugs those bugs are likely to be difficult to notice on
review. In addition, because these models were trained indiscriminately and
non-consensually on open-source code with a variety of licenses, it's not
obvious that we have the moral or legal right to redistribute code they
generate.