2022年以前的书籍
Pre-2022 Books

原始链接: https://notes.lorenzogravina.com/musings/pre-2022-books

作者观察到自己潜意识里更偏爱 2022 年之前出版的书籍,认为纯人工创作——包括起草、编辑和校对——所投入的精力赋予了作品一种当代 AI 辅助作品所缺乏的价值。 尽管作者意识到这种立场存在矛盾——鉴于他们自己也精通大语言模型,并清楚即便使用 AI 也能产出高质量成果——但他们仍难以将这种理性的认知与对传统写作中“人力成本”的感性珍视调和起来。作者将这种不安与历史上对印刷术或互联网等技术进步的质疑进行对比,承认自己可能只是在抵制一种必然的变革。最终,作者仍不确定这种偏好是某种固有的偏见,还是对不断演变的媒介所做出的正当反应,并认为随着时间的推移和熟悉程度的加深,这种内在冲突或许终将消解。

这篇 Hacker News 讨论探讨了人工智能崛起导致当代内容价值下降的现象。 一位用户表示对更新自己的技术书籍感到犹豫,担心现代的时间戳或与 AI 相关的编辑会剥离其“人类编写”的真实感。其他人则对 AI 生成内容的质量展开了辩论,指出虽然目前的产出往往陈词滥调且缺乏原创性,但它最终可能会像摄影术的早期阶段一样,发展成为一种合法的艺术媒介。 讨论随后转向了“内容垃圾”这一更广泛的问题。一些人认为,低质量 AI 生成内容的泛滥是现有媒体激励机制(特别是对“应用停留时间”指标的追求)的症状,而不仅仅是 AI 技术的副产品。最终,参与者们认为,算法优化内容的涌入,加速了长期以来困扰数字领域的问题。
相关文章

原文

I noticed that I seem to, subconsciously, gravitate towards books published on or before 2022, and somewhat discount books published after, especially from authors I haven’t heard of.

A part of me tells me I shouldn’t feel this. I like and use LLMs often for coding work, and I know that they can create great results. And after all, if the result is good, who cares about what tool was used to create it?

When I read a pre-2022 book, I know that each word was typed in manually, checked manually, edited manually, proofread manually. That, somehow, has an effect on me, and leads me to give greater weight to the book and what’s said in it.

I don’t want to sound like those who “worried” about society being dumbed down by writing, printing, newspapers, radio, television, and then the Internet. I’d just be yelling at the clouds. But I can’t help but feel that the effort meant something.

I don’t know what the solution to this is, and there doesn’t have to be one. Perhaps we’ll get used to this new tool, and move on.

联系我们 contact @ memedata.com