LLMs in Issues could lead to a "Talent deficit" ?

[senft-research]

So I have noticed there is, from my understanding, an LLM agent that handles the majority of the issues on the project now.

If that is the case, how do Maintainers expect new contributors to start working on the project? If the Issues are being resolved automatically, surely there are no practical ways to begin contributing to the project effectively?

I ask as I am an academic who has been looking into this problem with several FOSS projects that utilise LLMs, and trying to get an understanding of if Maintainers of such projects have identified this issue, and how they are mitigating it.

People's thoughts would be appreciated.