Ask HN：AI 副驾驶正在削弱我们的编程技能吗？
2 分•作者: buscoideais•6 个月前
在过去的12个月里，我将AI副驾驶（GitHub Copilot、Tabnine等）整合到了我的日常工作流程中。它们加速了样板代码的编写，建议单行修复方案，甚至可以根据需求重构整个函数。<p>但我注意到了一些令人不安的事情：<p>* <i>理解肤浅</i>：我有时会接受建议，但并没有完全理解它们。
* <i>解决问题的能力下降</i>：在解决难题时，我感觉独立找到解决方案的信心不足。
* <i>新开发人员的融入</i>：初级工程师依赖AI的输出，而不质疑边缘情况，这会导致一些不易察觉的错误。<p>向社区提出的问题：<p>* 自从采用AI工具以来，您是否经历过技能退化或责任感下降？
* 在利用AI速度的同时，哪些实践有助于您保持深入的理解？
* 我们应该将AI副驾驶视为“草稿生成器”还是真正的编程伙伴？<p>我很乐意听到轶事、策略或确凿的数据。让我们一起找出如何在不成为这些强大助手的学徒的情况下使用它们。
查看原文
Over the last 12 months I’ve integrated AI copilots (GitHub Copilot, Tabnine, etc.) into my daily workflow. They speed up boilerplate, suggest one-line fixes, and even refactor entire functions on demand.<p>But I’ve noticed something unsettling:<p>* <i>Shallow Understanding</i>: I sometimes accept suggestions without fully understanding them.
* <i>Problem-Solving Rust</i>: On hard problems, I feel less confident in reaching a solution independently.
* <i>Onboarding New Devs</i>: Junior engineers rely on AI outputs without questioning edge cases, leading to subtle bugs.<p>Questions for the community:<p>* Have you experienced skill atrophy or decreased ownership since adopting AI tools?
* What practices help you preserve deep understanding while still leveraging AI speed?
* Should we treat AI copilots as “draft generators” or as true programming partners?<p>I’d love to hear anecdotes, strategies, or hard data. Let’s figure out how to use these powerful assistants without becoming their apprentices.