AI Writing Is Killing Credibility, Even in Open Source

Two separate threads today flagged LLM-generated copy as a credibility problem. The OpenLogi project (an open-source Logitech driver replacement) got called out in the comments: 'I hate how every new site has this obviously LLM generated copy. Most people who use it for their copy don't realize how obvious it is.' The Turbovec vector search library got the same note: 'It would be nice to have the README be a little more human written for a project where you actually want people to adopt it.'

The pattern here is that LLM-generated text is now recognizable enough that it actively undermines trust in technical projects. For a developer tool or open-source library, the README is the first handshake with a potential contributor or user. When it reads like it was produced by a chatbot, the implicit message is that the author does not care enough to write it themselves, which raises questions about how much they care about everything else.

This is a narrower version of a broader credibility problem: the same signal that makes marketing copy feel cheap is now bleeding into technical documentation, grant proposals, and project announcements.