请问HN：在我们的人际沟通中平衡人工智能的使用
我们在工作流程中都在使用大型语言模型（LLM）这一事实不容否认。然而，当我们收到一封明显经过 LLM “润色”的电子邮件，或者在 HN 上看到一条显然是 LLM “改写”的评论时，心中难免会产生一种自动的挫败感。
为什么当有人做我们大家都在做的事情时，我们会感到不快？是因为明显的（过度）使用吗？
我已经到了不再像以前那样修订自己的写作的地步，为了保持“真实”，我故意保留一些小的拼写错误和语法瑕疵——这些在非母语英语使用者中是很常见的。
毫无疑问，我们将会面对越来越多的 AI 生成内容，而“我们”指的是我们人类及我们的代理人，问题是何时这种情况会达到平衡？我们何时才能接受这一事实？
这是否与我们愿意为“手工制作”商品支付额外费用有关？（作为类比：很明显，你的鞋子是机器制造的。）
（*）或许除了 99.999% 的人之外，大家都是如此。
——
为了讽刺，这里是上述内容的“由 Claude 润色”版本
——
问 HN：在人与人之间的沟通中平衡 AI 的使用
我们在工作流程中都在使用 LLM。这并不是一个有争议的说法。然而，当你收到一封明显经过 LLM “润色”的电子邮件，或阅读一条显然是 AI 改写的评论时，几乎所有人都会感到恼怒。这种挫败感是真实存在的，虚伪也是如此。
这是否因为它的明显性？缺乏努力？还是更令人不安的原因——我们认出这张面具，因为我们自己也在戴着它？
我已经到了故意不再修订自己写作的地步，保留一些小的拼写错误和语法怪癖。这在非母语者中是很常见的。我把这些作为存在的证明。
AI 生成内容的数量只会不断增加。不仅仅是人类对其做出选择——还有我们的代理人，代表我们与彼此的代理人沟通。挫败感何时会变得正常？我们何时会将其视为基础设施，就像我们接受电子邮件或自动更正一样？
在手工制作的商品中有一个有用的类比。我们知道鞋子是机器制造的，但我们仍然愿意为手工制作的鞋子支付更高的价格。这种溢价并不在于鞋子本身——而在于信号。证明一个人选择花时间去制作。
也许这就是我们真正反应的原因：并不是 AI 的帮助本身，而是缺乏这种信号。
查看原文
It is a fact that we are all* using LLM throughout our workflows.
And yet, there is an automatic frustration when one's receiving an email that was clearly "polished" by an LLM, or say, a comment here on HN that was "rephrased" by an LLM.<p>Why are we getting upset when someone is doing what we all do?
Is it the obvious (over) use?<p>It got to a point where I am not revising my writing as I used to, so it will remain "authentic" (leaving minor spelling and perhaps some grammatical errors - common for non native English speakers).<p>there is no questions we are going deal with growing amount of ai generated content, and by we it is us, humans, and our agents, and the question is when will it balanced out? when we will all accept it as a fact?<p>Is it related to the fact we are willing to pay extra for "hand crafted" goods?
(as an analogy: it is obvious that you had your machine generated my shoe)<p>(*) perhaps another all bu 99.999% of us.<p>--
For the irony in it, below is the "Polished by Claude" version of the above
--<p>ASK HN: Balancing AI use in human-to-human communication<p>We're all using LLMs throughout our workflows. That's not a controversial claim.
And yet there's a near-universal irritation when you receive an email that was clearly "polished" by one — or read a comment that was obviously AI-rephrased. The frustration is real, and so is the hypocrisy.<p>Is it the obviousness of it? The lack of effort? Or something more uncomfortable — that we recognize the mask because we wear it ourselves?
It's gotten to the point where I've stopped revising my own writing, deliberately leaving minor spelling errors and grammatical quirks intact. Common for non-native speakers. I keep them as proof of presence.<p>The volume of AI-generated content will only grow. Not just from humans making choices about it — but from our agents, acting on our behalf, communicating with each other's agents. At what point does the frustration normalize? When do we simply accept it as infrastructure, the way we accepted email or autocorrect?<p>There's a useful analogy in handcrafted goods. We know the shoe was machine-made. We still pay more for the one that wasn't. The premium isn't about the shoe — it's about the signal. Proof that a human chose to spend time.<p>Maybe that's what we're really reacting to: not the AI assistance itself, but the absence of the signal.