跳到正文
Hacker News:AI 熱帖·· 8 天前AI 評分35

拒絕 AI 低質內容:宣言

No AI Slop: A Manifesto

AI 導讀

討論引用宣言的觀點:AI 生成千行 PR 或十頁文件的成本正接近零,但審核成本仍然存在,有評論者認為審核負擔可能增加。有人建議在程式碼貢獻中附上提示詞,方便理解生成過程;亦有人指出,多輪提示詞、測試及評估流程難以用單一提示詞呈現。

正文

Image 1Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

No AI Slop: A Manifesto (slopmanifesto.com) 20 points by bankombinator36 minutes ago | hide | past | favorite | 4 comments help Image 2courtcircuits29 minutes ago | next[–] > Improving the system over patching the output This really connect with the latest conversations I had with coworkers regarding how can we review gigantic PR generated by AI. One way could be to ask the submitter to give their prompt as part of their contribution. This way it's easier for the reviewer to understand what was the thought process of the contributor and if this very process is sound. Not saying that this way to review is a silver bullet, but at least it removes some weight. reply Image 3viklove9 minutes ago | parent | next[–] Do most people generate PRs from a one-shot prompt? I have a whole pipeline of prompts, tests and evals, which loop back and potentially restart if the result is inadequate. How would I include that in the PR? reply Image 4jurgenburgen0 minutes ago | parent | prev | next[–] How does reviewing the long discussion the author had with the LLM prove the code is correct? reply Image 5mschuster912 minutes ago | prev[–] > Slop is asymmetric. The cost of producing a thousand-line PR or a ten-page doc is approaching zero, while the cost of reviewing it remains constant. I'd say the cost actually grows as people get laid off left and right, and on top of that, a 1000 line PR from a good human, give me a case of beer and I'll make me some sense of it, but if it's AI slop, get ready to hand me a crate of homemade Slivo or I won't touch that... reply Image 6 Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact Search:

來源:Hacker News:AI 熱帖 · news.ycombinator.com