中国对 OpenClaw （小龙虾）的需求正在推高二手 MacBook 的价格
正文翻译
OpenClaw demand in China is driving up the price of used MacBooks
中国对 OpenClaw 的需求正在推高二手 MacBook 的价格。
中国对 OpenClaw 的需求正在推高二手 MacBook 的价格。
评论翻译
Prime_1
likes: 14
Can someone explain why I would want this?
谁能解释一下我为什么会需要这东西？
AwwHellsNo
likes: 6
Niche reasons.
Maybe you like coding and want to be able to send a Whatsapp message to an Agent that then turns on/off appliances in your home for instance. This makes it easier.
一些小众的原因。
比如，也许你喜欢编程，并且希望能够向 AI 代理发送一条 WhatsApp 消息，然后让它帮你打开或关闭家里的电器。这东西能让这些操作变得更容易。
Lumbearjack
likes: 13
Ah, so we soundly don't need it.
啊，所以我们确实完全不需要它。
Lumbearjack
likes: -4
Thankfully no one "needs" this. It's a neat unreliable toy, but no one needs it. Every imagined use case is, "well, what if you're incompetent and want to do something pointless?"
谢天谢地，没有人“需要”这玩意。它就是个花哨但不可靠的玩具，但根本没人需要它。每一个设想的用例都是：“好吧，如果你能力不足又想干点毫无意义的事情该怎么办？”
Lumbearjack
likes: -4
Trusting a megacorporation"s dream machine to cater your news is sad as hell.
Tell every writer or actor who works in tv that we don't need tvs. Tell every artist we don't need entertainment. The Slop only aims to replace us and add nothing in return. And most of all, it's wrong. The more you know in your field the more wrong you see it is.
Plus it's not a product. It's a service pushed by your device maker who desperately need a way to profit off of their lost billions.
指望大公司的造梦机器来为你推送新闻，这真是悲哀透顶。
去告诉每一个在电视行业工作的编剧或演员我们不需要电视。去告诉每一个艺术家我们不需要娱乐。这种“电子垃圾”的目的只是取代我们，而不会提供任何实质性的回报。最重要的是，它错得离谱。你在自己的专业领域了解得越多，就越能发现它有多荒谬。
而且它根本不是一个产品。它是设备制造商强推的一项服务，因为他们迫切需要一种方法来弥补损失的数十亿美元并从中获利。
orangefilmgarden
likes: 2
Didn't we already have thousands of appliances that for some reason need to be connected to wifi, and allow me to do this already via apps, no AI necessary? My AC and even my rice cooker for some reason have these features, and I have had both appliances since 2020. Never used those features, though.
我们不是早就有成千上万种不知为何非要连 WiFi 的电器了吗？而且不用 AI，我早就通过各种 App 实现了这些功能吧？我的空调甚至电饭煲不知为何都有这些功能，我从 2020 年起就买了这俩电器。不过，我从来没用过那些功能。
SeaFox2142
likes: 1
Oh comme on, don't tell me you don't want to inject more money and data for all those cool billionaires doing wonders around the world???!!!
噢得了吧，别告诉我你不想给那些在世界各地创造奇迹的酷炫亿万富翁们注入更多的资金和数据？？？！！！
Lowetheiy
likes: 1
1995: internet demand drives up the price of used PCs
Luddite: Can someone explain why I would want this?
1995 年：对互联网的需求推高了二手个人电脑的价格。
卢德分子（反对技术进步者）：谁能解释一下我为什么会需要这东西？
Calcutec_1
likes: 133
“OpenClaw is an AI agent, a tool that can autonomously conduct personal tasks such as sending emails and shopping online. “
How does that even make sense ? You have tell it what to buy or what to say in the email right ? So you might as well do it yourself
“OpenClaw 是一款 AI 代理，一种可以自主执行发送电子邮件和在线购物等个人任务的工具。”
这根本说不通好吗？你还是得告诉它要买什么，或者在邮件里写什么，对吧？既然如此，你还不如自己动手。
MetaRecruiter
likes: 98
It’s pretty much a psyop. It’s a way for hopeful business owners and developers to embark on a fruitless journey of burning tokens and making ai providers more money.
这差不多就是一种心理战。这只是一种让充满希望的企业主和开发者踏上燃烧代币的徒劳之旅的方式，好让人工智能提供商赚更多的钱。
mrwafu
likes: 45
It’s absolutely baffling that people are diving head first into it when we literally saw it go crazy and dexe a high level Facebook manager’s email against her wishes
https://www.pcmag.com/news/meta-security-researchers-openclaw-ai-agent-accidentally-dexed-her-emails
当大家明明眼睁睁看着这玩意发疯，违背一名 Facebook 高管的意愿删除了她的电子邮件，人们居然还对它趋之若鹜，这简直让人无法理解。
Red-Star-44
likes: 32
My CEO yesterday showed me a video of the nvidia CEO showcasing their openclaw version (nemoclaw) and he was saying shit like how this nemoclaw will be used by every company in the world, how you need to have a strategy for it and how it will change the world, dude was speaking like a cult leader and my CEO just takes what he says as facts for some reason.
昨天我们 CEO 给我看了一段英伟达 CEO 展示他们自家的 OpenClaw 版本（Nemoclaw）的视频。他在那瞎扯什么世界上每家公司都将使用这个 Nemoclaw，你必须为它制定战略，它将如何改变世界等等。这家伙说话简直就像个邪教头子，而我们 CEO 竟然不知怎么的就把他说的当成了真理。
neverfakemaplesyrup
likes: 1
It genuinely is a cult at this point.
I work at a ski shop. Way too many high end clients are obsessed with AI but when I ask what it actually does for them they just draw a blank then get frustrated because it has to be worthwhile, right? They were told it's valuable, so it has to be! One just pulled out his phone, put in "Tell me why you are useful to me exactly", and smugly showed me the resulting slop.
I recently became underemployed and went to Salesforce in yet another vain attempt to break into a better career. The last I saw it, it was more or less a lot of business jargon. Now all the regular stuff- Salesforce architect, salesforce admin, etc is just gone, it's all "training" that is just "GUYS PLEASE WE SWEAR OUR AI IS GOOD AND VALUABLE PLEASE PLEASE PLEASE USE IT FOR THE LOVE OF GOD"
发展到这个地步，它真的已经成邪教了。
我在一家滑雪用品店工作。有太多高端客户对 AI 痴迷不已，但当我问他们 AI 到底能为他们做些什么时，他们却大脑一片空白，然后变得十分烦躁。毕竟这东西必须得有价值才对，是吧？别人告诉他们这很有价值，所以它必须得有！有个人直接掏出手机输入：“确切地告诉我，你到底对我有什么用”，然后得意洋洋地把生成的那堆电子垃圾展示给我看。
我最近处于半失业状态，跑去尝试 Salesforce 的职位，试图再次徒劳地挤进一个更好的职业赛道。上次我了解的时候，它差不多还只是一堆商业术语。而现在，所有常规的岗位——Salesforce 架构师、Salesforce 管理员等等全都不见了，剩下的全是所谓的“培训”，骨子里全是在喊叫：“大家伙儿求求你们了，我们发誓我们的 AI 非常好非常有价值，看在上帝的份上求求你们快用它吧！”
pimpeachment
likes: -46
Anyone who works in leadership understands the high impact Ai is currently having and the obvious expected ramp up. Leaders need to be accounting for the risk introduced by these tools and their future unknown risks, financially, operationally, technically, and regulatory. When the ceo of the most valuable company says something is happening, believe them until proven otherwise. You might not like it, but reality of Ai in business is here. If you don't adopt Ai, you will be replaced by someone who did adopt it.
任何从事领导工作的人都明白 AI 目前所产生的高影响力以及显然可预见的迅猛增长。领导者必须考虑到这些工具带来的风险及其未来的未知风险，包括财务、运营、技术和监管层面。当最有价值公司的首席执行官断言某件事正在发生时，在被证明是错误之前，你最好相信他们。你可能不喜欢这样，但 AI 融入商业的现实已经摆在眼前。如果你不接受 AI，你就会被接受 AI 的人所取代。
I-Am-Maldoror
likes: 42
> When the ceo of the most valuable company says something is happening, believe them until proven otherwise.
Yeah, like when Zuckerberg was talking about metaverse. That is so great.
> 当最有价值公司的首席执行官断言某件事正在发生时，在被证明是错误之前，你最好相信他们。
是啊，就像扎克伯格当时大谈元宇宙的时候一样。那可真是太棒了。
pimpeachment
likes: -34
Zuckerberg isn't wrong. A metaverse style future for Internet will happen. The scale and timing are unknown and his version of it failed. It doesn't mean it won't happen. VR failed many times before it became useful. AI will have many failures and many successes in the future.
扎克伯格并没有错。互联网的未来必然会出现一种类似元宇宙的形态。虽然具体规模和时机尚不确定，而且他那个版本确实失败了，但并不意味着这不会发生。VR 在变得实用之前也经历了多次失败。在未来，AI 同样会经历许多失败，也会取得许多成功。
toorigged2fail
likes: 11
Are you writing this from your VR room at home?
你是在你家的 VR 房间里写下这些话的吗？
Weekly-Fortune2611
likes: 17
CEOs are the last people you should be listening to. Their job is pump up stock price anyway they can. They will do anything within legal limits for it ( and sometimes cross legal limits )
CEO 是你最不应该听信的一群人。他们的工作就是想尽一切办法推高股价。为了达到这个目的，他们会在法律允许的范围内做任何事（有时甚至会跨越法律红线）。
pimpeachment
likes: 5
I understand your reasoning and in a rational world you are right. However, the USA is an oligarchy and the top CEOs control the direction of the economy. You should trust what they say because it impacts you and the entire economy when they talk. AI agents will take over tasks. It already happening. It doesn't matter if we think it's "slop" it matters that it is in fact happening already. Accept the new reality or suffer wishing for the past to return.
我理解你的逻辑，在一个理性的世界里，你是对的。然而，美国是个寡头政治社会，顶级的 CEO 们控制着经济的走向。你应该相信他们所说的话，因为当他们发话时，就会对你和整个经济产生影响。AI 代理将接管各种任务。这已经成为现实。无论我们是否认为它是“电子垃圾”都不重要，重要的是它确确实实已经发生了。要么接受新的现实，要么就痛苦地期盼着回到过去吧。
Weekly-Fortune2611
likes: 6
Oh I don’t doubt executives will direct their companies to implant half baked hallucination prone ai solutions where they are not needed, in order to appear innovative and justify their compensation and promos
哦，我毫不怀疑高管们会指示他们的公司，在根本不需要的地方强行植入那些半生不熟、容易产生幻觉的 AI 解决方案，好让自己显得很有创新精神，并借此为自己的高薪和升职进行辩护。
anarchyx34
likes: 4
I run an instance on my homelab server. I’m still figuring it out so I don’t have it doing much yet but my philosophy will be to treat it like an employee, meaning it gets its own everything. It’s own email, own GitHub account, etc. It has mostly free rein to do whatever it wants on it’s own instance short of root access, but my files it accesses through shared cloud storage folders from it’s own account, same as you’d do with any other coworker. I don’t expect it to run my life or an entire startup for me but I do see the potential for it to solve actual problems for me and perform helpful daily tasks that would be difficult to accomplish with anything else, and it already has. I plan on giving it it’s own iCloud account too so it we can share things with each other without giving it access to mine.
My biggest issue with it so far is that it’s really temperamental and most of the time I’ve spent with it is getting it to fix itself. Just look at the repo, there’s over 5k open PR’s. Right now to me it’s a fun, and sometimes very clever toy that I can make do some useful things.
我在自家的家庭实验室服务器上运行了一个实例。我还在摸索阶段，所以还没让它做太多事，但我的理念是把它当成员工来对待，这意味着它拥有属于自己的一切：专属的电子邮件、专属的 GitHub 账户等等。在它的专属实例上，只要没有 Root 权限，它基本上可以自由地做任何它想做的事情，但当它访问我的文件时，是通过它的个人账户访问共享云存储文件夹进行的，就像你和其他同事协作时那样。
我没指望它能替我打理整个生活或者运营一家初创公司，但我确实看到了它有潜力为我解决实际问题，并执行一些用其他工具难以完成的日常有益任务，实际上它已经做到了一些。我打算也给它配备一个专属的 iCloud 账户，这样我们就可以彼此共享内容，同时又无需授予它访问我私人账户的权限。
目前我遇到最大的问题是它“脾气很古怪”，我花在它身上的大部分时间都是在让它自我修复。去代码仓库查看就会发现，那里有超过 5000 个未解决的拉取请求（PR）。对我来说，现在它只是个好玩、偶尔很聪明的玩具，我能让它干些有用的事。
arika_ex
likes: 1
"I don’t expect it to run my life or an entire startup for me but I do see the potential for it to solve actual problems for me and perform helpful daily tasks that would be difficult to accomplish with anything else, and it already has."
Could you give some examples? Of course just broadly speaking is fine if some things are sensitive.
“我没指望它能替我打理整个生活或者运营一家初创公司，但我确实看到了它有潜力为我解决实际问题，并执行一些用其他工具难以完成的日常有益任务，实际上它已经做到了一些。”
你能举几个例子吗？当然，如果涉及敏感信息，泛泛而谈也是可以的。
Pjpjpjpjpj
likes: 15
That is a super simplistic descxtion of what it does.
China is pushing AI and automation very hard, at the national and local levels, so everyone is encouraged to use it, including funding to help them set up. Major tech companies are pushing OpenClaw, so there is a ton of advertising. And there is a real viral community built around it, with a feeling that "everyone else is doing it and I'm losing out if I don't.
Among overworked white collar workers, people selling online, analysts, and people in finance, there is belief that it can help them automate pieces of their work, allowing them to be more productive and thus either have more time or advance more quickly. There is a big following in the entrepreneur space, with people believing this will enable them to operate the illustrious "one person company" to escape the rat race, or achieve prosperity.
Some things it can do:
Ecommerce: Automate product listings (e.g. provide a spreadsheet of data for 50 products and it will automatically add them into your account on multiple online sales marketplaces. Great for small operations or individual trying to make a few bucks online.
Ecommerce: Automate tracking competitor prices and provide price recommendations for all your products. Great for small operations or individual trying to make a few bucks online.
Product Marketing: Continuously scrape competitor websites for new product launches and automate recommendations for a company to react (change pricing, change features, change packaging, launch new competitive product, etc.).
Social Media Communications: Automatically create SEO social media content, schedules their release, and posts them across a variety of platforms. This can be reactive (e.g. competitor launches new product, it generates social media response and automatically blasts the internet with a 30-day campaign).
Finance: Automate crypto currency trades based upon internet triggers (e.g. if an executive posts the phrase X, then buy/sell).
Investment: Automated analysis of SEC-type filings instantly, immediately updating valuation models and providing stock buy/sell recommendations instantly. This is work many people do full time, and a tool that can allow someone to be 5x as productive while also giving results in 1 minute rather than 1 day is appealing to people in the occupation.
A big example is the "Sanwan" combination of 8 individual agents. People see it as a means for creating a "one person" company - key for people looking to escape the rat race, or to escape poverty in more rural areas. It can crank out AI-generated content allowing an individual with no real tech skills to generate engagement on their website for ecommerce, influencer, affiliate marketing, or whatever purposes.
Maybe it will all fizzle out, but it is far more than just helping people shop online.
这只是对它功能超级简化的描述。
中国在国家和地方层面上都在大力推进 AI 和自动化，所以鼓励每个人去使用它，甚至提供资金帮助他们进行部署。各大科技公司都在强推 OpenClaw，因此广告铺天盖地。围绕着它还建立起了一个极具病毒式传播效应的真实社区，给人一种“别人都在用，我不用就会吃亏”的氛围感。
在那些过度劳累的白领、电商卖家、分析师以及金融从业者群体中，大家相信 AI 能帮助他们自动化处理部分工作，从而提高生产力，进而获得更多休息时间或者晋升得更快。在创业圈子里也有一大批追随者，人们相信这能让他们打造出大名鼎鼎的“一人公司”，从而逃避内卷，或是走向致富之路。
它能做的一些事情包括：
电子商务：自动化产品上架（例如，提供一份包含 50 种产品数据的电子表格，它会自动将这些产品添加到你各大在线销售平台的账户中。非常适合小本经营或想在网上赚点外快的个人）。
电子商务：自动追踪竞争对手的价格，并为你所有的产品提供定价建议。非常适合小本经营或想在网上赚点外快的个人。
产品营销：持续抓取竞争对手的网站以获取新产品发布信息，并自动为企业提供应对建议（例如调整定价、修改功能、更换包装、推出新的对标产品等）。
社交媒体传播：自动生成符合 SEO（搜索引擎优化）的社媒内容，安排发布计划，并自动发布到各大平台。这甚至可以是响应式的（例如，当竞争对手发布新产品时，它会自动生成社媒回应，并发起为期 30 天的全网轰炸式营销）。
金融：根据互联网上的触发条件自动执行加密货币交易（例如，如果某位高管发布了短语 X，它就立刻执行买入/卖出）。
投资：立刻自动分析诸如 SEC（美国证券交易委员会）发布的公告文件，立即更新估值模型并瞬间提供股票买卖建议。很多人的全职工作就是做这个，如果有一种工具能让某人的工作效率提升 5 倍，并且在 1 分钟内就能给出原本需要 1 天才能算出的结果，那这对业内人士来说绝对是极具吸引力的。
一个突出的例子是由 8 个独立 AI 代理组成的“三湾（Sanwan）”组合。人们将其视为创建“一人公司”的工具——这对于那些渴望逃避内卷，或者试图在偏远农村地区摆脱贫困的人来说至关重要。它能批量生成 AI 内容，让一个毫无真正技术背景的个人也能在自己的网站上吸引流量，从而实现电商销售、网红带货、分销联盟推广等各种目的。
也许这一切最终都会像泡沫一样破灭，但它能做的事情绝不仅仅是帮你网购那么简单。
catgirl-lover-69
likes: 12
Redditors hate AI and LLMs. At work I am taking every system management task that I find mundane or boring and automating with AI. And yeah there’s a risk it might blow the entire thing up, who cares though
Reddit 网友非常痛恨 AI 和大语言模型（LLM）。在工作中，我把每一个我觉得枯燥乏味的系统管理任务都交给了 AI 去自动化处理。的确，这存在着把一切都搞砸的风险，但谁在乎呢。
lafadeaway
likes: 13
Companies that handle sensitive data should definitely care. The biggest problems with LLMs continue to be security and hallucinations.
Want to spin up OpenClaw to organize your personal calendar? Go ahead.
But use OpenClaw to file your client’s taxes? Fuck no.
处理敏感数据的公司绝对应该在乎。大语言模型最大的问题依然是安全漏洞和“幻觉”。
想启动 OpenClaw 来帮你整理私人日程安排？去吧。
但想用 OpenClaw 帮你客户报税？去他妈的，绝对不行。
SantaCruzHostel
likes: 3
I tried to have Claude give me estimates for splitting a road trip into equal legs and help find things to do along the route. It kept trying to tell me one leg was ~500 miles when Google Maps shows it over 900 miles. Ai seems helpful when the answer is subjective (help me rewrite this email to sound more professional) or when you already know the answer, but it seems to fail miserably when trying to provide hard facts.
我曾试图让 Claude 帮我估算如何将一段公路旅行平均分成几段，并帮我找沿途有哪些好玩的地方。它一直试图告诉我其中一段路程大约是 500 英里，然而谷歌地图上明明显示这一段超过 900 英里。当答案是主观的时候（比如：帮我重写这封邮件，让它听起来更专业），或者当你已经知道答案的时候，AI 似乎很有用；但是当试图让它提供确凿的事实数据时，它似乎败得一塌糊涂。
anarchyx34
likes: 2
Tell it to use a python scxt to do the calculation and it will get the correct answer. LLM’s don’t do math. They’re *language* models.
告诉它写个 Python 脚本来进行计算，它就能得出正确的答案。大语言模型并不擅长做数学运算。它们是“语言”模型。
Randromeda2172
likes: -1
This sounds like a skill issue. You found the one program that's not built to do math, and you told it to add numbers. That's not what its for. A Porsche isn't going to handle offroading, doesn't mean that the car is a waste.
听起来这像是个使用者“技术不够”的问题。你找了一个并非为数学计算而构建的程序，却偏让它去加数字。那不是它的用途。保时捷无法胜任越野跑道，但这并不意味着那辆车就是废物。
Middle-Nerve1732
likes: 9
You could just say “find me the best deal on a [thing] and order one.” It will present you with a bunch of options then you approve the one you want and it orders for you. It’s like having a full time assistant to do any tasks you might have done yourself in front of your laptop.
你可以直接吩咐它：“帮我找一款性价比最高的【某件商品】，然后买下来。” 它会为你呈现一堆选项，然后你只要批准你想要的那个，它就会替你下单。这就像雇了一个全职助理，能代替你完成任何以往你在笔记本电脑前亲自做的事情。
Calcutec_1
likes: 12
That sounds like the scammers dream tbh,
说实话，这听起来简直就是诈骗犯梦寐以求的东西。
phranq
likes: 4
So if I ask it to find me a doctor who is in network within 5 miles with certain criteria and is accepting new patients and make me an appointment it definitely isn’t going to need to have its hand held the entire time right? Because I could also just do that myself if it can’t.
那么，如果我让它帮我在医保网络内、5 英里范围内找一个符合某些特定条件，且正在接受新患者的医生，并帮我预约，它肯定不需要我全程手把手教它对吧？因为如果它做不到，我也完全可以自己去搞定。
AstariiFilms
likes: 1
You can prompt it to gather all applicable doctors offices in a 5 mile radius, load a voice model, have it call each office, ask and transcribe each response, and filter the results based on your criteria.
你可以给它下指令，让它收集方圆 5 英里内所有符合要求的医生诊所，加载一个语音模型，让它给每家诊所打电话询问并转录所有的回复，最后根据你的标准过滤筛选出结果。
robotnique
likes: 3
Oh God. That's the last thing we need and will quickly devolve into AIs at each Doctors office having a weird handshake with your AI assistant as they decide your care and then you and the doctor show up for your appointment and only then actually figure out what any of this was for.
天啊。这恰恰是我们最不需要的，而且很快就会演变成这样一种滑稽的场面：每个医生诊所的 AI 与你的私人 AI 助手进行某种奇葩的“握手”协议，由它们来决定你的护理方案。然后，只有当你和医生真正面对面出现在诊室里时，你们俩才终于搞清楚这一切到底是为什么。
TeslasAndComicbooks
likes: 1
I looked into it to see why it was so damn hard for me to find a Mac Mini. It’s worth looking into. It’s pretty cool. Obviously not for everyone but can streamline mundane tasks pretty easily and depending on what industry you’re in, run operations while you sleep, etc…
I personally have no use for it but it’s very early.
我专门调查了一下，想看看为什么我买台 Mac Mini 怎么就那么费劲。这东西还是值得研究的。它挺酷的。显然它并不适合所有人，但它可以非常轻松地简化那些繁杂单调的日常任务，而且根据你所处的行业，它甚至能在你睡觉的时候替你处理业务，等等……
我个人是用不上它，不过这项技术现在还处于非常早期的阶段。
Gisschace
likes: 1
Reading some of the lengths they go to do really basic tasks; like booking Ubers, makes me realise it’s not worth it yet.
Like how many Ubers is someone booking that this saves them time? Then there’s the cost of the tools. Takes me 30 seconds to book an uber, I do it while I am putting on my coat and leaving the house.
看到他们为了完成像叫个 Uber 这样极其基础的任务，居然要费那么大周折，让我意识到目前这玩意根本不值得。
就比如说，一个人要叫多少辆 Uber 才能真正让他觉得这东西省了时间？更别提这些工具还有成本。我自己叫个 Uber 只要 30 秒，我通常是一边穿外套准备出门一边就叫了。
e1epi
likes: -2
This is what I don't understand about "AI" agents like anything it can do I can do faster and with less stress than getting one of them to do it and what I want it to do I either can't trust it to do or if I want advice on lets say stocks, it won't give a summarized idea on anything financial related.
The stuff is practically worthless imo.
这就是我对所谓的“AI 代理”最无法理解的地方。它能做的任何事情，我自己做起来都比指望它去做要快得多，而且压力更小；而那些我真正希望它去做的事情，我要么不敢相信它能做好，要么比如我想寻求关于股票的建议，它又拒绝提供任何与金融相关的总结性观点。
在我看来，这些东西几乎就是一文不值的废物。
SeaFox2142
likes: 1
Like someone said "if you don't know about the subject, you can't trust AI. If you know, you have to double check it's work. Why are they useful? (Besides replying to emails and creating text)"
正如有人说的那样：“如果你不懂那个领域，你就不能信任 AI。如果你懂，你又必须得反复检查它的工作。那么它们到底有什么用呢？（除了回回邮件和写写文章之外）”
Infinite_Maximum_820
likes: -6
Why don’t you use it and learn yourself ?
你为什么不亲自用用，自己去了解一下呢？
Calcutec_1
likes: -1
Because i have literally no use for it ? What possible use would a normal citizen get out of this ?
因为我真的用不上它啊？一个普通老百姓能用它干点啥？
Infinite_Maximum_820
likes: 0
Anything you can do with a computer it can.. if you do nothing productive with a computer, then sure. Otherwise people usually work with a computer and want to get shit done faster
你能用电脑做的任何事，它都能做……如果你用电脑从来不干正事，那确实。不然的话，人们通常会用电脑工作，而且总是希望能更快地把活儿干完。
Calcutec_1
likes: 5
But if i tell it what to do, and then review and approve or reject what it does, then i can just as well do it myself.
但如果我必须得吩咐它怎么做，然后再去审查、批准或拒绝它做出的成果，那我还不如直接自己干算了。
Infinite_Maximum_820
likes: 2
No, you’re being dumb. Try it out and learn why thousands of people are doing it without anyone forcing them to
不，你这想法太蠢了。亲自去试一试吧，弄明白为什么有成千上万的人在没人强迫的情况下，依然自发地在使用它。
Calcutec_1
likes: -1
Not until someone explains a good use case for an average person
除非有人能给出一个对普通人来说很棒的使用案例，否则我才不用。
phranq
likes: 1
This is the first technology to that I can remember where I’m told it can do “everything” and when I ask to see it do this “magic” I get the constant run around.
这是我记忆中第一项被称为“无所不能”的技术，然而当我要求看它施展这种“魔法”时，得到的却总是顾左右而言他的含糊其辞。
AstariiFilms
likes: 5
There's tons of examples in these comments, people see them and hand wave them away because "I can do that too" without realizing that they no longer have to do that thing at all. how much time does that open up for other things? It's like seeing a washing machine and saying "I can wash clothes too".
这些评论里有大量的例子，但人们看到后总是一笑而过，因为他们觉得“这种事我自己也能做”，却没有意识到他们其实已经完全不需要再亲自动手去做那件事了。这能为你腾出多少时间去做其他的事情？这就像你看到一台洗衣机，然后轻描淡写地说：“我自己也会洗衣服啊。”
Calcutec_1
likes: 13
So spam, this is going to be used to spam. Great, we need more of that.
所以就是发垃圾信息呗，这玩意绝对会被用来疯狂发送垃圾信息。好极了，我们这世界可太需要更多垃圾信息了。
thatguygreg
likes: 29
The first thing I read about OpenClaw was that not only was it not a good idea to be used on corp devices, but if you had installed it, it should be reported as a security incident.
How the F has this thing gotten anywhere?
我看到的关于 OpenClaw 的第一条消息就是：它不仅不适合在公司设备上使用，而且如果你已经安装了它，那就应该将其上报为一起安全事件。
这特么到底是怎么流行起来的？
AndroidUser37
likes: 11
Well, that's the reason everyone's buying used machines. If you install it on a dedicated computer, and keep your valuable information away from it, you can give it free reign with less to worry about. Fixes the security risk.
嗯，这就是大家都在抢购二手机器的原因。如果你把它安装在一台专用电脑上，并让它远离你所有的重要信息，你就可以毫无顾虑地让它自由发挥了。这也从根本上解决了安全风险问题。
IAmCorgii
likes: 15
Except without valuable information, what are you doing with it? If the thing doesn't have, for example, my payment accounts or calendar or anything, what is it doing for me?
问题是，如果不给它提供这些高价值的信息，那你用它来干嘛呢？如果这玩意连我的支付账户或者日程日历什么都没有，它到底还能替我干些啥？
Eze-Wong
likes: 7
It's popular in the US too. A lot of people are buying up mac minis to run their own OpenClaws. I don't get it honestly because it opens up too much vulnerability and if you don't meter your tokens properly you can end up doing something like, having two AI agents talk to each other and wasting tokens. happened to a guy who spent thousands.
它在美国也很火。很多人都在抢购 Mac Mini 来运行他们自己的 OpenClaw。老实说我不太理解，因为这会暴露太多的安全漏洞。而且如果你没有正确管控好你 API 调用时消耗的代币（tokens），你最后可能会搞出那种“让两个 AI 代理互相对话”的乌龙，从而白白浪费掉大量代币。有个倒霉蛋就因为这事损失了好几千美元。
DZello
likes: 5
I’m surprised to see that people are so busy that they can’t even find time to shop online themselves.
我很惊讶地发现，人们竟然已经忙到连自己上网买个东西的时间都抽不出来了。
rotoboro
likes: -2
No idea where you got the idea that’s all they’re using it for.
真不知道你从哪得出的结论，认为他们用这玩意儿就仅仅只是为了网购。
robotnique
likes: 8
OpenClaw, find some witty response to this list and make sure I come across as really smart but not overly smug.
OpenClaw，帮我想一条风趣幽默的回复来回应上面这一长串内容，并确保让我显得非常聪明，但又不要过于自鸣得意。
likes: 14
Can someone explain why I would want this?
谁能解释一下我为什么会需要这东西？
AwwHellsNo
likes: 6
Niche reasons.
Maybe you like coding and want to be able to send a Whatsapp message to an Agent that then turns on/off appliances in your home for instance. This makes it easier.
一些小众的原因。
比如，也许你喜欢编程，并且希望能够向 AI 代理发送一条 WhatsApp 消息，然后让它帮你打开或关闭家里的电器。这东西能让这些操作变得更容易。
Lumbearjack
likes: 13
Ah, so we soundly don't need it.
啊，所以我们确实完全不需要它。
Lumbearjack
likes: -4
Thankfully no one "needs" this. It's a neat unreliable toy, but no one needs it. Every imagined use case is, "well, what if you're incompetent and want to do something pointless?"
谢天谢地，没有人“需要”这玩意。它就是个花哨但不可靠的玩具，但根本没人需要它。每一个设想的用例都是：“好吧，如果你能力不足又想干点毫无意义的事情该怎么办？”
Lumbearjack
likes: -4
Trusting a megacorporation"s dream machine to cater your news is sad as hell.
Tell every writer or actor who works in tv that we don't need tvs. Tell every artist we don't need entertainment. The Slop only aims to replace us and add nothing in return. And most of all, it's wrong. The more you know in your field the more wrong you see it is.
Plus it's not a product. It's a service pushed by your device maker who desperately need a way to profit off of their lost billions.
指望大公司的造梦机器来为你推送新闻，这真是悲哀透顶。
去告诉每一个在电视行业工作的编剧或演员我们不需要电视。去告诉每一个艺术家我们不需要娱乐。这种“电子垃圾”的目的只是取代我们，而不会提供任何实质性的回报。最重要的是，它错得离谱。你在自己的专业领域了解得越多，就越能发现它有多荒谬。
而且它根本不是一个产品。它是设备制造商强推的一项服务，因为他们迫切需要一种方法来弥补损失的数十亿美元并从中获利。
orangefilmgarden
likes: 2
Didn't we already have thousands of appliances that for some reason need to be connected to wifi, and allow me to do this already via apps, no AI necessary? My AC and even my rice cooker for some reason have these features, and I have had both appliances since 2020. Never used those features, though.
我们不是早就有成千上万种不知为何非要连 WiFi 的电器了吗？而且不用 AI，我早就通过各种 App 实现了这些功能吧？我的空调甚至电饭煲不知为何都有这些功能，我从 2020 年起就买了这俩电器。不过，我从来没用过那些功能。
SeaFox2142
likes: 1
Oh comme on, don't tell me you don't want to inject more money and data for all those cool billionaires doing wonders around the world???!!!
噢得了吧，别告诉我你不想给那些在世界各地创造奇迹的酷炫亿万富翁们注入更多的资金和数据？？？！！！
Lowetheiy
likes: 1
1995: internet demand drives up the price of used PCs
Luddite: Can someone explain why I would want this?
1995 年：对互联网的需求推高了二手个人电脑的价格。
卢德分子（反对技术进步者）：谁能解释一下我为什么会需要这东西？
Calcutec_1
likes: 133
“OpenClaw is an AI agent, a tool that can autonomously conduct personal tasks such as sending emails and shopping online. “
How does that even make sense ? You have tell it what to buy or what to say in the email right ? So you might as well do it yourself
“OpenClaw 是一款 AI 代理，一种可以自主执行发送电子邮件和在线购物等个人任务的工具。”
这根本说不通好吗？你还是得告诉它要买什么，或者在邮件里写什么，对吧？既然如此，你还不如自己动手。
MetaRecruiter
likes: 98
It’s pretty much a psyop. It’s a way for hopeful business owners and developers to embark on a fruitless journey of burning tokens and making ai providers more money.
这差不多就是一种心理战。这只是一种让充满希望的企业主和开发者踏上燃烧代币的徒劳之旅的方式，好让人工智能提供商赚更多的钱。
mrwafu
likes: 45
It’s absolutely baffling that people are diving head first into it when we literally saw it go crazy and dexe a high level Facebook manager’s email against her wishes
https://www.pcmag.com/news/meta-security-researchers-openclaw-ai-agent-accidentally-dexed-her-emails
当大家明明眼睁睁看着这玩意发疯，违背一名 Facebook 高管的意愿删除了她的电子邮件，人们居然还对它趋之若鹜，这简直让人无法理解。
Red-Star-44
likes: 32
My CEO yesterday showed me a video of the nvidia CEO showcasing their openclaw version (nemoclaw) and he was saying shit like how this nemoclaw will be used by every company in the world, how you need to have a strategy for it and how it will change the world, dude was speaking like a cult leader and my CEO just takes what he says as facts for some reason.
昨天我们 CEO 给我看了一段英伟达 CEO 展示他们自家的 OpenClaw 版本（Nemoclaw）的视频。他在那瞎扯什么世界上每家公司都将使用这个 Nemoclaw，你必须为它制定战略，它将如何改变世界等等。这家伙说话简直就像个邪教头子，而我们 CEO 竟然不知怎么的就把他说的当成了真理。
neverfakemaplesyrup
likes: 1
It genuinely is a cult at this point.
I work at a ski shop. Way too many high end clients are obsessed with AI but when I ask what it actually does for them they just draw a blank then get frustrated because it has to be worthwhile, right? They were told it's valuable, so it has to be! One just pulled out his phone, put in "Tell me why you are useful to me exactly", and smugly showed me the resulting slop.
I recently became underemployed and went to Salesforce in yet another vain attempt to break into a better career. The last I saw it, it was more or less a lot of business jargon. Now all the regular stuff- Salesforce architect, salesforce admin, etc is just gone, it's all "training" that is just "GUYS PLEASE WE SWEAR OUR AI IS GOOD AND VALUABLE PLEASE PLEASE PLEASE USE IT FOR THE LOVE OF GOD"
发展到这个地步，它真的已经成邪教了。
我在一家滑雪用品店工作。有太多高端客户对 AI 痴迷不已，但当我问他们 AI 到底能为他们做些什么时，他们却大脑一片空白，然后变得十分烦躁。毕竟这东西必须得有价值才对，是吧？别人告诉他们这很有价值，所以它必须得有！有个人直接掏出手机输入：“确切地告诉我，你到底对我有什么用”，然后得意洋洋地把生成的那堆电子垃圾展示给我看。
我最近处于半失业状态，跑去尝试 Salesforce 的职位，试图再次徒劳地挤进一个更好的职业赛道。上次我了解的时候，它差不多还只是一堆商业术语。而现在，所有常规的岗位——Salesforce 架构师、Salesforce 管理员等等全都不见了，剩下的全是所谓的“培训”，骨子里全是在喊叫：“大家伙儿求求你们了，我们发誓我们的 AI 非常好非常有价值，看在上帝的份上求求你们快用它吧！”
pimpeachment
likes: -46
Anyone who works in leadership understands the high impact Ai is currently having and the obvious expected ramp up. Leaders need to be accounting for the risk introduced by these tools and their future unknown risks, financially, operationally, technically, and regulatory. When the ceo of the most valuable company says something is happening, believe them until proven otherwise. You might not like it, but reality of Ai in business is here. If you don't adopt Ai, you will be replaced by someone who did adopt it.
任何从事领导工作的人都明白 AI 目前所产生的高影响力以及显然可预见的迅猛增长。领导者必须考虑到这些工具带来的风险及其未来的未知风险，包括财务、运营、技术和监管层面。当最有价值公司的首席执行官断言某件事正在发生时，在被证明是错误之前，你最好相信他们。你可能不喜欢这样，但 AI 融入商业的现实已经摆在眼前。如果你不接受 AI，你就会被接受 AI 的人所取代。
I-Am-Maldoror
likes: 42
> When the ceo of the most valuable company says something is happening, believe them until proven otherwise.
Yeah, like when Zuckerberg was talking about metaverse. That is so great.
> 当最有价值公司的首席执行官断言某件事正在发生时，在被证明是错误之前，你最好相信他们。
是啊，就像扎克伯格当时大谈元宇宙的时候一样。那可真是太棒了。
pimpeachment
likes: -34
Zuckerberg isn't wrong. A metaverse style future for Internet will happen. The scale and timing are unknown and his version of it failed. It doesn't mean it won't happen. VR failed many times before it became useful. AI will have many failures and many successes in the future.
扎克伯格并没有错。互联网的未来必然会出现一种类似元宇宙的形态。虽然具体规模和时机尚不确定，而且他那个版本确实失败了，但并不意味着这不会发生。VR 在变得实用之前也经历了多次失败。在未来，AI 同样会经历许多失败，也会取得许多成功。
toorigged2fail
likes: 11
Are you writing this from your VR room at home?
你是在你家的 VR 房间里写下这些话的吗？
Weekly-Fortune2611
likes: 17
CEOs are the last people you should be listening to. Their job is pump up stock price anyway they can. They will do anything within legal limits for it ( and sometimes cross legal limits )
CEO 是你最不应该听信的一群人。他们的工作就是想尽一切办法推高股价。为了达到这个目的，他们会在法律允许的范围内做任何事（有时甚至会跨越法律红线）。
pimpeachment
likes: 5
I understand your reasoning and in a rational world you are right. However, the USA is an oligarchy and the top CEOs control the direction of the economy. You should trust what they say because it impacts you and the entire economy when they talk. AI agents will take over tasks. It already happening. It doesn't matter if we think it's "slop" it matters that it is in fact happening already. Accept the new reality or suffer wishing for the past to return.
我理解你的逻辑，在一个理性的世界里，你是对的。然而，美国是个寡头政治社会，顶级的 CEO 们控制着经济的走向。你应该相信他们所说的话，因为当他们发话时，就会对你和整个经济产生影响。AI 代理将接管各种任务。这已经成为现实。无论我们是否认为它是“电子垃圾”都不重要，重要的是它确确实实已经发生了。要么接受新的现实，要么就痛苦地期盼着回到过去吧。
Weekly-Fortune2611
likes: 6
Oh I don’t doubt executives will direct their companies to implant half baked hallucination prone ai solutions where they are not needed, in order to appear innovative and justify their compensation and promos
哦，我毫不怀疑高管们会指示他们的公司，在根本不需要的地方强行植入那些半生不熟、容易产生幻觉的 AI 解决方案，好让自己显得很有创新精神，并借此为自己的高薪和升职进行辩护。
anarchyx34
likes: 4
I run an instance on my homelab server. I’m still figuring it out so I don’t have it doing much yet but my philosophy will be to treat it like an employee, meaning it gets its own everything. It’s own email, own GitHub account, etc. It has mostly free rein to do whatever it wants on it’s own instance short of root access, but my files it accesses through shared cloud storage folders from it’s own account, same as you’d do with any other coworker. I don’t expect it to run my life or an entire startup for me but I do see the potential for it to solve actual problems for me and perform helpful daily tasks that would be difficult to accomplish with anything else, and it already has. I plan on giving it it’s own iCloud account too so it we can share things with each other without giving it access to mine.
My biggest issue with it so far is that it’s really temperamental and most of the time I’ve spent with it is getting it to fix itself. Just look at the repo, there’s over 5k open PR’s. Right now to me it’s a fun, and sometimes very clever toy that I can make do some useful things.
我在自家的家庭实验室服务器上运行了一个实例。我还在摸索阶段，所以还没让它做太多事，但我的理念是把它当成员工来对待，这意味着它拥有属于自己的一切：专属的电子邮件、专属的 GitHub 账户等等。在它的专属实例上，只要没有 Root 权限，它基本上可以自由地做任何它想做的事情，但当它访问我的文件时，是通过它的个人账户访问共享云存储文件夹进行的，就像你和其他同事协作时那样。
我没指望它能替我打理整个生活或者运营一家初创公司，但我确实看到了它有潜力为我解决实际问题，并执行一些用其他工具难以完成的日常有益任务，实际上它已经做到了一些。我打算也给它配备一个专属的 iCloud 账户，这样我们就可以彼此共享内容，同时又无需授予它访问我私人账户的权限。
目前我遇到最大的问题是它“脾气很古怪”，我花在它身上的大部分时间都是在让它自我修复。去代码仓库查看就会发现，那里有超过 5000 个未解决的拉取请求（PR）。对我来说，现在它只是个好玩、偶尔很聪明的玩具，我能让它干些有用的事。
arika_ex
likes: 1
"I don’t expect it to run my life or an entire startup for me but I do see the potential for it to solve actual problems for me and perform helpful daily tasks that would be difficult to accomplish with anything else, and it already has."
Could you give some examples? Of course just broadly speaking is fine if some things are sensitive.
“我没指望它能替我打理整个生活或者运营一家初创公司，但我确实看到了它有潜力为我解决实际问题，并执行一些用其他工具难以完成的日常有益任务，实际上它已经做到了一些。”
你能举几个例子吗？当然，如果涉及敏感信息，泛泛而谈也是可以的。
Pjpjpjpjpj
likes: 15
That is a super simplistic descxtion of what it does.
China is pushing AI and automation very hard, at the national and local levels, so everyone is encouraged to use it, including funding to help them set up. Major tech companies are pushing OpenClaw, so there is a ton of advertising. And there is a real viral community built around it, with a feeling that "everyone else is doing it and I'm losing out if I don't.
Among overworked white collar workers, people selling online, analysts, and people in finance, there is belief that it can help them automate pieces of their work, allowing them to be more productive and thus either have more time or advance more quickly. There is a big following in the entrepreneur space, with people believing this will enable them to operate the illustrious "one person company" to escape the rat race, or achieve prosperity.
Some things it can do:
Ecommerce: Automate product listings (e.g. provide a spreadsheet of data for 50 products and it will automatically add them into your account on multiple online sales marketplaces. Great for small operations or individual trying to make a few bucks online.
Ecommerce: Automate tracking competitor prices and provide price recommendations for all your products. Great for small operations or individual trying to make a few bucks online.
Product Marketing: Continuously scrape competitor websites for new product launches and automate recommendations for a company to react (change pricing, change features, change packaging, launch new competitive product, etc.).
Social Media Communications: Automatically create SEO social media content, schedules their release, and posts them across a variety of platforms. This can be reactive (e.g. competitor launches new product, it generates social media response and automatically blasts the internet with a 30-day campaign).
Finance: Automate crypto currency trades based upon internet triggers (e.g. if an executive posts the phrase X, then buy/sell).
Investment: Automated analysis of SEC-type filings instantly, immediately updating valuation models and providing stock buy/sell recommendations instantly. This is work many people do full time, and a tool that can allow someone to be 5x as productive while also giving results in 1 minute rather than 1 day is appealing to people in the occupation.
A big example is the "Sanwan" combination of 8 individual agents. People see it as a means for creating a "one person" company - key for people looking to escape the rat race, or to escape poverty in more rural areas. It can crank out AI-generated content allowing an individual with no real tech skills to generate engagement on their website for ecommerce, influencer, affiliate marketing, or whatever purposes.
Maybe it will all fizzle out, but it is far more than just helping people shop online.
这只是对它功能超级简化的描述。
中国在国家和地方层面上都在大力推进 AI 和自动化，所以鼓励每个人去使用它，甚至提供资金帮助他们进行部署。各大科技公司都在强推 OpenClaw，因此广告铺天盖地。围绕着它还建立起了一个极具病毒式传播效应的真实社区，给人一种“别人都在用，我不用就会吃亏”的氛围感。
在那些过度劳累的白领、电商卖家、分析师以及金融从业者群体中，大家相信 AI 能帮助他们自动化处理部分工作，从而提高生产力，进而获得更多休息时间或者晋升得更快。在创业圈子里也有一大批追随者，人们相信这能让他们打造出大名鼎鼎的“一人公司”，从而逃避内卷，或是走向致富之路。
它能做的一些事情包括：
电子商务：自动化产品上架（例如，提供一份包含 50 种产品数据的电子表格，它会自动将这些产品添加到你各大在线销售平台的账户中。非常适合小本经营或想在网上赚点外快的个人）。
电子商务：自动追踪竞争对手的价格，并为你所有的产品提供定价建议。非常适合小本经营或想在网上赚点外快的个人。
产品营销：持续抓取竞争对手的网站以获取新产品发布信息，并自动为企业提供应对建议（例如调整定价、修改功能、更换包装、推出新的对标产品等）。
社交媒体传播：自动生成符合 SEO（搜索引擎优化）的社媒内容，安排发布计划，并自动发布到各大平台。这甚至可以是响应式的（例如，当竞争对手发布新产品时，它会自动生成社媒回应，并发起为期 30 天的全网轰炸式营销）。
金融：根据互联网上的触发条件自动执行加密货币交易（例如，如果某位高管发布了短语 X，它就立刻执行买入/卖出）。
投资：立刻自动分析诸如 SEC（美国证券交易委员会）发布的公告文件，立即更新估值模型并瞬间提供股票买卖建议。很多人的全职工作就是做这个，如果有一种工具能让某人的工作效率提升 5 倍，并且在 1 分钟内就能给出原本需要 1 天才能算出的结果，那这对业内人士来说绝对是极具吸引力的。
一个突出的例子是由 8 个独立 AI 代理组成的“三湾（Sanwan）”组合。人们将其视为创建“一人公司”的工具——这对于那些渴望逃避内卷，或者试图在偏远农村地区摆脱贫困的人来说至关重要。它能批量生成 AI 内容，让一个毫无真正技术背景的个人也能在自己的网站上吸引流量，从而实现电商销售、网红带货、分销联盟推广等各种目的。
也许这一切最终都会像泡沫一样破灭，但它能做的事情绝不仅仅是帮你网购那么简单。
catgirl-lover-69
likes: 12
Redditors hate AI and LLMs. At work I am taking every system management task that I find mundane or boring and automating with AI. And yeah there’s a risk it might blow the entire thing up, who cares though
Reddit 网友非常痛恨 AI 和大语言模型（LLM）。在工作中，我把每一个我觉得枯燥乏味的系统管理任务都交给了 AI 去自动化处理。的确，这存在着把一切都搞砸的风险，但谁在乎呢。
lafadeaway
likes: 13
Companies that handle sensitive data should definitely care. The biggest problems with LLMs continue to be security and hallucinations.
Want to spin up OpenClaw to organize your personal calendar? Go ahead.
But use OpenClaw to file your client’s taxes? Fuck no.
处理敏感数据的公司绝对应该在乎。大语言模型最大的问题依然是安全漏洞和“幻觉”。
想启动 OpenClaw 来帮你整理私人日程安排？去吧。
但想用 OpenClaw 帮你客户报税？去他妈的，绝对不行。
SantaCruzHostel
likes: 3
I tried to have Claude give me estimates for splitting a road trip into equal legs and help find things to do along the route. It kept trying to tell me one leg was ~500 miles when Google Maps shows it over 900 miles. Ai seems helpful when the answer is subjective (help me rewrite this email to sound more professional) or when you already know the answer, but it seems to fail miserably when trying to provide hard facts.
我曾试图让 Claude 帮我估算如何将一段公路旅行平均分成几段，并帮我找沿途有哪些好玩的地方。它一直试图告诉我其中一段路程大约是 500 英里，然而谷歌地图上明明显示这一段超过 900 英里。当答案是主观的时候（比如：帮我重写这封邮件，让它听起来更专业），或者当你已经知道答案的时候，AI 似乎很有用；但是当试图让它提供确凿的事实数据时，它似乎败得一塌糊涂。
anarchyx34
likes: 2
Tell it to use a python scxt to do the calculation and it will get the correct answer. LLM’s don’t do math. They’re *language* models.
告诉它写个 Python 脚本来进行计算，它就能得出正确的答案。大语言模型并不擅长做数学运算。它们是“语言”模型。
Randromeda2172
likes: -1
This sounds like a skill issue. You found the one program that's not built to do math, and you told it to add numbers. That's not what its for. A Porsche isn't going to handle offroading, doesn't mean that the car is a waste.
听起来这像是个使用者“技术不够”的问题。你找了一个并非为数学计算而构建的程序，却偏让它去加数字。那不是它的用途。保时捷无法胜任越野跑道，但这并不意味着那辆车就是废物。
Middle-Nerve1732
likes: 9
You could just say “find me the best deal on a [thing] and order one.” It will present you with a bunch of options then you approve the one you want and it orders for you. It’s like having a full time assistant to do any tasks you might have done yourself in front of your laptop.
你可以直接吩咐它：“帮我找一款性价比最高的【某件商品】，然后买下来。” 它会为你呈现一堆选项，然后你只要批准你想要的那个，它就会替你下单。这就像雇了一个全职助理，能代替你完成任何以往你在笔记本电脑前亲自做的事情。
Calcutec_1
likes: 12
That sounds like the scammers dream tbh,
说实话，这听起来简直就是诈骗犯梦寐以求的东西。
phranq
likes: 4
So if I ask it to find me a doctor who is in network within 5 miles with certain criteria and is accepting new patients and make me an appointment it definitely isn’t going to need to have its hand held the entire time right? Because I could also just do that myself if it can’t.
那么，如果我让它帮我在医保网络内、5 英里范围内找一个符合某些特定条件，且正在接受新患者的医生，并帮我预约，它肯定不需要我全程手把手教它对吧？因为如果它做不到，我也完全可以自己去搞定。
AstariiFilms
likes: 1
You can prompt it to gather all applicable doctors offices in a 5 mile radius, load a voice model, have it call each office, ask and transcribe each response, and filter the results based on your criteria.
你可以给它下指令，让它收集方圆 5 英里内所有符合要求的医生诊所，加载一个语音模型，让它给每家诊所打电话询问并转录所有的回复，最后根据你的标准过滤筛选出结果。
robotnique
likes: 3
Oh God. That's the last thing we need and will quickly devolve into AIs at each Doctors office having a weird handshake with your AI assistant as they decide your care and then you and the doctor show up for your appointment and only then actually figure out what any of this was for.
天啊。这恰恰是我们最不需要的，而且很快就会演变成这样一种滑稽的场面：每个医生诊所的 AI 与你的私人 AI 助手进行某种奇葩的“握手”协议，由它们来决定你的护理方案。然后，只有当你和医生真正面对面出现在诊室里时，你们俩才终于搞清楚这一切到底是为什么。
TeslasAndComicbooks
likes: 1
I looked into it to see why it was so damn hard for me to find a Mac Mini. It’s worth looking into. It’s pretty cool. Obviously not for everyone but can streamline mundane tasks pretty easily and depending on what industry you’re in, run operations while you sleep, etc…
I personally have no use for it but it’s very early.
我专门调查了一下，想看看为什么我买台 Mac Mini 怎么就那么费劲。这东西还是值得研究的。它挺酷的。显然它并不适合所有人，但它可以非常轻松地简化那些繁杂单调的日常任务，而且根据你所处的行业，它甚至能在你睡觉的时候替你处理业务，等等……
我个人是用不上它，不过这项技术现在还处于非常早期的阶段。
Gisschace
likes: 1
Reading some of the lengths they go to do really basic tasks; like booking Ubers, makes me realise it’s not worth it yet.
Like how many Ubers is someone booking that this saves them time? Then there’s the cost of the tools. Takes me 30 seconds to book an uber, I do it while I am putting on my coat and leaving the house.
看到他们为了完成像叫个 Uber 这样极其基础的任务，居然要费那么大周折，让我意识到目前这玩意根本不值得。
就比如说，一个人要叫多少辆 Uber 才能真正让他觉得这东西省了时间？更别提这些工具还有成本。我自己叫个 Uber 只要 30 秒，我通常是一边穿外套准备出门一边就叫了。
e1epi
likes: -2
This is what I don't understand about "AI" agents like anything it can do I can do faster and with less stress than getting one of them to do it and what I want it to do I either can't trust it to do or if I want advice on lets say stocks, it won't give a summarized idea on anything financial related.
The stuff is practically worthless imo.
这就是我对所谓的“AI 代理”最无法理解的地方。它能做的任何事情，我自己做起来都比指望它去做要快得多，而且压力更小；而那些我真正希望它去做的事情，我要么不敢相信它能做好，要么比如我想寻求关于股票的建议，它又拒绝提供任何与金融相关的总结性观点。
在我看来，这些东西几乎就是一文不值的废物。
SeaFox2142
likes: 1
Like someone said "if you don't know about the subject, you can't trust AI. If you know, you have to double check it's work. Why are they useful? (Besides replying to emails and creating text)"
正如有人说的那样：“如果你不懂那个领域，你就不能信任 AI。如果你懂，你又必须得反复检查它的工作。那么它们到底有什么用呢？（除了回回邮件和写写文章之外）”
Infinite_Maximum_820
likes: -6
Why don’t you use it and learn yourself ?
你为什么不亲自用用，自己去了解一下呢？
Calcutec_1
likes: -1
Because i have literally no use for it ? What possible use would a normal citizen get out of this ?
因为我真的用不上它啊？一个普通老百姓能用它干点啥？
Infinite_Maximum_820
likes: 0
Anything you can do with a computer it can.. if you do nothing productive with a computer, then sure. Otherwise people usually work with a computer and want to get shit done faster
你能用电脑做的任何事，它都能做……如果你用电脑从来不干正事，那确实。不然的话，人们通常会用电脑工作，而且总是希望能更快地把活儿干完。
Calcutec_1
likes: 5
But if i tell it what to do, and then review and approve or reject what it does, then i can just as well do it myself.
但如果我必须得吩咐它怎么做，然后再去审查、批准或拒绝它做出的成果，那我还不如直接自己干算了。
Infinite_Maximum_820
likes: 2
No, you’re being dumb. Try it out and learn why thousands of people are doing it without anyone forcing them to
不，你这想法太蠢了。亲自去试一试吧，弄明白为什么有成千上万的人在没人强迫的情况下，依然自发地在使用它。
Calcutec_1
likes: -1
Not until someone explains a good use case for an average person
除非有人能给出一个对普通人来说很棒的使用案例，否则我才不用。
phranq
likes: 1
This is the first technology to that I can remember where I’m told it can do “everything” and when I ask to see it do this “magic” I get the constant run around.
这是我记忆中第一项被称为“无所不能”的技术，然而当我要求看它施展这种“魔法”时，得到的却总是顾左右而言他的含糊其辞。
AstariiFilms
likes: 5
There's tons of examples in these comments, people see them and hand wave them away because "I can do that too" without realizing that they no longer have to do that thing at all. how much time does that open up for other things? It's like seeing a washing machine and saying "I can wash clothes too".
这些评论里有大量的例子，但人们看到后总是一笑而过，因为他们觉得“这种事我自己也能做”，却没有意识到他们其实已经完全不需要再亲自动手去做那件事了。这能为你腾出多少时间去做其他的事情？这就像你看到一台洗衣机，然后轻描淡写地说：“我自己也会洗衣服啊。”
Calcutec_1
likes: 13
So spam, this is going to be used to spam. Great, we need more of that.
所以就是发垃圾信息呗，这玩意绝对会被用来疯狂发送垃圾信息。好极了，我们这世界可太需要更多垃圾信息了。
thatguygreg
likes: 29
The first thing I read about OpenClaw was that not only was it not a good idea to be used on corp devices, but if you had installed it, it should be reported as a security incident.
How the F has this thing gotten anywhere?
我看到的关于 OpenClaw 的第一条消息就是：它不仅不适合在公司设备上使用，而且如果你已经安装了它，那就应该将其上报为一起安全事件。
这特么到底是怎么流行起来的？
AndroidUser37
likes: 11
Well, that's the reason everyone's buying used machines. If you install it on a dedicated computer, and keep your valuable information away from it, you can give it free reign with less to worry about. Fixes the security risk.
嗯，这就是大家都在抢购二手机器的原因。如果你把它安装在一台专用电脑上，并让它远离你所有的重要信息，你就可以毫无顾虑地让它自由发挥了。这也从根本上解决了安全风险问题。
IAmCorgii
likes: 15
Except without valuable information, what are you doing with it? If the thing doesn't have, for example, my payment accounts or calendar or anything, what is it doing for me?
问题是，如果不给它提供这些高价值的信息，那你用它来干嘛呢？如果这玩意连我的支付账户或者日程日历什么都没有，它到底还能替我干些啥？
Eze-Wong
likes: 7
It's popular in the US too. A lot of people are buying up mac minis to run their own OpenClaws. I don't get it honestly because it opens up too much vulnerability and if you don't meter your tokens properly you can end up doing something like, having two AI agents talk to each other and wasting tokens. happened to a guy who spent thousands.
它在美国也很火。很多人都在抢购 Mac Mini 来运行他们自己的 OpenClaw。老实说我不太理解，因为这会暴露太多的安全漏洞。而且如果你没有正确管控好你 API 调用时消耗的代币（tokens），你最后可能会搞出那种“让两个 AI 代理互相对话”的乌龙，从而白白浪费掉大量代币。有个倒霉蛋就因为这事损失了好几千美元。
DZello
likes: 5
I’m surprised to see that people are so busy that they can’t even find time to shop online themselves.
我很惊讶地发现，人们竟然已经忙到连自己上网买个东西的时间都抽不出来了。
rotoboro
likes: -2
No idea where you got the idea that’s all they’re using it for.
真不知道你从哪得出的结论，认为他们用这玩意儿就仅仅只是为了网购。
robotnique
likes: 8
OpenClaw, find some witty response to this list and make sure I come across as really smart but not overly smug.
OpenClaw，帮我想一条风趣幽默的回复来回应上面这一长串内容，并确保让我显得非常聪明，但又不要过于自鸣得意。
