User talk:Garkeith

Click here to leave me a new message.

If I leave a message on your talk page, please respond here so the conversation stays in one place. I will generally respond to messages left on this page. Older conversations posted before July 28, 2026, have been moved to the talk page archives for organization and remain available through the archive links and page history. Thank you for your message and for contributing to Wikipedia.

September 2026

You have been blocked indefinitely from editing for resuming LLM use against unblock conditions and policy, as you did at Virgil L. Highland.

If you believe that there are good reasons for being unblocked, please review Wikipedia's guide to appealing blocks, then submit a request through the unblock wizard or through your talk page.  — rsjaffe 🗣️ 10:54, 10 September 2026 (UTC)

I have reviewed the current LLM rules and guidance carefully because I took the condition of my previous unblock seriously. My understanding has been that WP:LLM prohibits using an LLM to generate or rewrite article content, but that it does not prohibit every use of an LLM while researching or working on Wikipedia.

Wikipedia:Large language models specifically says that the ban does not cover indirect use of LLMs adjacent to editing. It gives examples including identifying structural problems in articles, generating ideas, asking for outlines, critiquing existing content, and suggesting possible improvements. Wikipedia:Responsibly using large language models also specifically discusses using AI tools to find possible sources. It says that the editor should manually read those sources and write the content themselves rather than relying on the LLM-written summary.

That is the distinction I believed I was following. I have continued to use AI as a research tool, including helping locate possible sources, identify areas that may need additional research, and examine existing articles for possible problems or missing material. I have not treated AI output as a source. When AI has identified a source that I may want to use, I have made an effort to open and review that source myself before relying on it.

I have also been going back through pages where I used LLM assistance in the past and trying to correct problems I may have introduced. I believed I was doing what the current guidance permits rather than violating my unblock condition. I was really trying to correct my past mistakes and If you review my edit history since you reinstated me you would easily see where I took really bad examples of LLM that I introduced and reworte them significantly.

The block reason points me to the Virgil L. Highland article, but I do not see an explanation on my talk page identifying what specific edit, passage, citation, or conduct in that article was considered a violation of the condition. WP:BLOCK says that administrators must supply a clear and specific reason why a user was blocked. Simply identifying an article does not tell me what I supposedly did wrong or what use of AI was considered prohibited.

I am not trying to argue about something before I understand the accusation. I am asking you to identify the specific edit or conduct that caused you to conclude that I had resumed prohibited LLM use, because based on my reading of the current rules I believed I was staying on the permitted research side of that line. One thing I can promise you, I have not been directly copying and pasting anything from AI like I was in the past before our last interaction. I think that should be obvious in my edits. Garkeith (talk) 04:17, 13 September 2026 (UTC)