{"slug": "human-voice", "title": "★ human voice", "summary": "A developer reports that LLM-generated issue reports are causing maintainers to auto-close or ignore legitimate bug reports, even when written in a human voice, leading him to pivot to using LLMs for issue creation with a follow-up human comment. The developer cites a Pi issue that was auto-closed despite following guidelines, and a fast-resume issue where he added a human-voice comment after an LLM-generated report.", "body_md": "♠ More and more I’ve been resorting to LLMs / Gen AI to file feature requests or issues/bugs in upstream repositories.\n\nSometimes I get completely ignored, even when I follow the project guidelines.\n\nFor example, last week I filed [this\nissue](https://github.com/earendil-works/pi/issues/7070) in\n[Pi](https://pi.dev). It got auto-closed. I know\n[why](https://github.com/earendil-works/pi/blob/main/CONTRIBUTING.md) it was\nauto-closed, but it’s lame, because I followed their guidelines. In spite of\nwriting it in my own voice, it got lost in the sea of slop issues. I can’t blame\nthe maintainers, I totally understandand and sympathize with them. Yet, it sucks\nfor me 1.\n\nIn other words: even when I put in the effort to file an issue manually, without using an LLM, it gets mistaken for an LLM slop issue, because that’s what everybody else is doing.\n\nWhich kind of incentives does this create?\n\n→ There’s little reason to put in the effort to write issues in my own voice.\n\nWhich is, ironically, the opposite of what the project maintainers seemingly hoped to accomplish (=to attract non-clanker contributors that file high-quality issues).\n\nSo now I am pivoting.\n\nI’ll shamelessly create issues (or feature requests) with an LLM. I’ll review\nthe wording first, but won’t attempt to write them in my own voice. Would it\neven matter? Maintainers are too busy and overwhelmed to be able to distinguish\nslop from human voice, which is ironical — LLMs were supposed to *improve* WLB,\nnot to make it worse. Oh well.\n\nThat said, pure Gen AI is lame. So I am starting to complement with a follow-up comment in my own voice.\n\nToday’s example: [angristan/fast-resume#87](https://github.com/angristan/fast-resume/issues/87#issuecomment-5343487168):\n\n(human voice)\n\nDisclaimer: the previous comment was created by an LLM, I reviewed its text though. I can reliably reproduce this locally. I’ve been using`fr`\n\n[for a while]now, it’s very handy. This is the only use case that is getting on my way.Let me know if you would like help to implement/address it. Thanks!\n\nCheck out [fast-resume](https://perrotta.dev/2026/05/fast-resume-search-coding-agent-sessions/) if you haven’t\nyet. It’s awesome.\n\nI think this is a good compromise. ∎\n\nAnd, frankly, for them too. I no longer feel motivated to file issues for\nPi, even though I’m becoming a heavy user of it. [↩︎](https://perrotta.dev/2026/08/human-voice/#fnref:1)\n\n— § —\n\nReply via [email](mailto:serendipity@perrotta.dev?subject=Reply to: human voice)", "url": "https://wpnews.pro/news/human-voice", "canonical_source": "https://perrotta.dev/2026/08/human-voice/", "published_at": "2026-08-19 14:24:06+00:00", "updated_at": "2026-08-19 15:12:32.112710+00:00", "lang": "en", "topics": ["generative-ai", "ai-ethics", "developer-tools"], "entities": ["Pi", "fast-resume", "angristan", "GitHub"], "alternates": {"html": "https://wpnews.pro/news/human-voice", "markdown": "https://wpnews.pro/news/human-voice.md", "text": "https://wpnews.pro/news/human-voice.txt", "jsonld": "https://wpnews.pro/news/human-voice.jsonld"}}