“AI氛围编码”是否在真实公司内部使原型设计变得更糟？
我不断听到有人说原型设计现在已经“解决”了——只需使用 Cursor、Claude、Lovable 等工具。
但当我与真实组织（如医疗、受监管行业，甚至大型非科技公司）中的人交谈时，我看到的却恰恰相反：
创意并不缺乏。人们希望测试的项目总是排着队——新的工作流程、内部工具、面向患者的流程、决策支持用户界面。
瓶颈不在于创造力，而在于：
- 专注于维护的内部IT团队
- 已经超负荷的工程师
- 仍然需要时间、上下文和责任感的AI工具
- 对于“仅仅是一个原型”来说，速度太慢或负担过重的代理机构/自由职业者
我的观点是：AI并没有消除原型设计的问题——它只是将这个问题转移到了那些最没有时间去处理它的人身上。
我很好奇这与你的经历是否相符：
- 你实际上是持续进行原型设计，还是大多数情况下只是一次性完成？
- AI工具是否完全取代了你对外部帮助的需求？
- 如果你能在几天内（而不是几个月）获得现实的原型，你会多频繁地使用这个机会？
我真心想了解我所看到的是否是一个真实的模式——还是仅仅是一个有偏见的视角。
查看原文
I keep hearing that prototyping is “solved” now — just use Cursor, Claude, Lovable, etc.<p>But when I talk to people inside real organizations (healthcare, regulated industries, even large non-tech companies), I keep seeing the opposite:<p>There’s no shortage of ideas. There’s a constant backlog of things people want to test — new workflows, internal tools, patient-facing flows, decision support UIs.<p>The bottleneck isn’t creativity. It’s:
– internal IT teams focused on maintenance
– engineers already overloaded
– AI tools that still require time, context, and ownership
– agencies/freelancers that are too slow or heavyweight for “just a prototype”<p>My hot take: AI didn’t eliminate the prototyping problem — it shifted it to the people who have the least time to deal with it.<p>Curious how this matches your experience:
– Do you actually prototype continuously, or is it mostly one-off?
– Have AI tools fully replaced the need for external help for you?
– If you could get realistic prototypes in days (not months), how often would you use that?<p>Genuinely trying to understand whether I’m seeing a real pattern — or just a biased slice of the world.