Meta 下架在自家门口用 Meta 眼镜拍的批评视频——HN 用户怎么看
原文概要
HN 热门榜上一个标题很直白的链接：Meta takes down a critical video about meta AI Glasses after filming at Meta（reddit.com），330 多分，170 多条评论。事情起源于荷兰讽刺视频创作者 Roel Maalderink——他做了十年讽刺题材（加沙、乌克兰、难民），过去视频从没被删过。这次他戴着 Meta 与雷朋合作的智能眼镜，到阿姆斯特丹 Meta 办公室门口对员工做街头采访，员工的五官被模糊处理、声音保留，发布到 Facebook/Instagram 后被 Meta 下架。
他本人直接引述：”I’ve been making satirical videos about all kinds of things for ten years: Gaza, Ukraine, refugees. Funny videos that push boundaries and provoke discussion. But I’ve never experienced a video being removed before.”——这句话被 HN 用户反复引用，作为”双重标准”的证据。HN 讨论由此迅速发酵到四个方向：法律管辖、隐私与 AI 训练、平台属性、街头骚扰边界。
视频原始链接（YouTube dHKhYL3is5o）目前在 YouTube 仍可观看，部分 HN 用户反馈遭遇了自动 AI 配音，体验混乱。
讨论焦点
双重标准：他们受不了被自家产品反向拍摄
“This hypocrisy is not at all surprising. The modus operandi for these people has been rules for thee, but not for me for quite a while now.”
— ElProlactin [c:49827967]
（译文：这虚伪不意外。这些人一直以来的套路就是「规则给你，但不给我」。）
ElProlactin 一锤定音后，HN 用户把同一指控展开成多种类比。最完整的一版来自 suttontom：
“It’s incredibly hypocritical. It’s like working at Flock and complaining about being filmed on a Flock camera. Or working for a company that sells cars that emit a loud, annoying sound and complaining about those cars in your neighborhood. Or making devices for children but banning your own children from using them because the devices are harmful.”
— suttontom [c:49829373]
（译文：极度虚伪。就像在 Flock 公司上班却抱怨被 Flock 摄像头拍到；就像卖发出噪音的车却抱怨邻居的车吵；就像做儿童产品却不让自己孩子用，因为有害。）
而 tjpnz 直接给眼镜起了个外号：
“Epstein Glasses by Meta. For people who don’t do consent.”
— tjpnz [c:49828511]
（译文：Meta 出的「爱泼斯坦眼镜」，专给不在乎 consent 的人用。）
63stack 把争论从法律层面拉回事实：法律说什么不重要，重要的是 Meta 员工对自己造的工具有多不自在：
“The hypocrisy is that meta is creating a tool that their employees are very uncomfortable seeing used on them. What the law says about consent is irrelevant to the hypocrisy, and the videos are automatically uploaded to meta without anyone’s consent anyway.”
— 63stack [c:49829124]
（译文：虚伪的点在于——Meta 造了这个工具，他们自己的员工被拍到就很不安。法律说 consent 怎么算跟虚伪无关，何况视频本来就在员工不知情的情况下自动上传给了 Meta。）
mns 进一步举例，Meta 内容审核长期”装死”：
“From Meta to OpenAi, for a lot of people here it’s just a ‘yeah, but actually’. I’ve reported everything from holocaust glorification to videos of people being filmed in public in europe just to be humiliated, to obvious far right propaganda done by russian linked accounts, nothing goes against Meta’s policy anymore, except when it’s about them…”
— mns [c:49829058]
（译文：从 Meta 到 OpenAI，这里很多人就是「是的，但是实际上」。我举报过大屠杀美化、欧洲街头羞辱被偷拍、明显的俄罗斯账号做的极右宣传——没一条撞 Meta 政策，除非被举报的是 Meta 自己。）
公开录像的法律拼图：荷兰合法、德国违法、葡萄牙特殊
一个 HN 用户认为视频里没拍到别人不愿被拍。Almondsetat 反驳”不，这侵犯了 consent”；ElProlactin 反驳”荷兰法律允许”——这条线展开成欧美各国法律的拼图。
“It is perfectly legal in the Netherlands to record people in public places. You don’t need someone’s consent to record in public, and if any of the employees objected to the publication of the recording, they could consult with an attorney about a portrait rights claim. Such a claim should be adjudicated by a court, not Facebook.”
— ElProlactin [c:49828097]
（译文：荷兰法律完全允许在公共场所拍人。公开拍摄不需要取得对方同意。如果被拍到的员工对发布有异议，可以走律师肖像权诉讼。这种诉求应由法院裁定，不该由 Facebook 来判。）
ElProlactin 后面还引用了具体法条：
“Under Article 139a of the Dutch Criminal Code, you are legally allowed to record any conversation in which you are a participant. You don’t need to inform or obtain permission from another party.”
— ElProlactin [c:49828228]
（译文：依据《荷兰刑法典》第 139a 条，作为对话参与者，录制该对话是合法的。你不需要告知或取得对方许可。）
但 Vinnl 提醒一个事实——员工的脸已经被模糊：
“Note that the employees were not recognisable; their faces were blurred out.”
— Vinnl [c:49828457]
（译文：注意，员工的脸不可辨识，已经打了马赛克。）
Barrin92 仍然认为肖像权（GDPR）足以支持 Meta 的下架：
“It’s perfectly legal to record but publish, especially in but not limited to a commercial context without explicit consent violates portrait rights under the GDPR in the EU.”
— Barrin92 [c:49828281]
（译文：录制合法，但发布——尤其在商业语境下未经明确同意——违反欧盟 GDPR 下的肖像权。）
德国用户 lionkor 直接说”德国不允许”：
“It’s illegal to record people in public spaces in Germany.”
— lionkor [c:49828115]
（译文：在德国公共场所拍人是违法的。）
但这条也立刻被 martin_a 部分修正——德国并非一刀切，重点在”个人是否成为画面主体”：
“It’s perfectly fine to take pictures of the Brandenburger Tor for example and post them online. There will likely be people in that picture but as they are not the main aspect of the image, it will be fine. It’s something different if you zoom in on somebody in that scene and he becomes the main point of the image.”
— martin_a [c:49828737]
（译文：拍勃兰登堡门、里面有路人，完全没问题——他们不是画面主体。一旦你放大某个人使其成为画面主体，个人权利就压过记录公共空间的权利。）
这一圈下来，HN 用户意识到一个问题：欧盟内部对同一行为的合法性判断都不一样——荷兰允许、德国因历史原因限制（bonoboTP [c:49828348] 提到”这也解释了德国当初抵制 Google Street View”）、爱尔兰公开场所无隐私期待、rithdmc [c:49828151] 说”在爱尔兰公共场所拍人不构成问题”，GDPR 又在商业发布层面再加一层。ElProlactin 由此反问 Meta：既然在德国可能违法，Meta 怎么还在德国卖？
“if this would be against the law in Germany, why is Facebook waiting for these glasses to be banned? It never should have made these available in Germany, and should be using technical means to ensure that they cannot function inside of Germany’s borders. So it’s just more proof of the hypocrisy. Meta doesn’t care about the privacy of the plebs.”
— ElProlactin [c:49828233]
（译文：如果德国算违法，Facebook 干嘛要等别人禁？它本来就不该在德国卖这款眼镜，或者应该用技术手段让它在德国境内根本不能工作。这进一步证明虚伪——Meta 不在乎老百姓的隐私。）
隐私与 AI 训练：你的视频可能正在 Kenya 被审
pj_mukh 一开始想护 Meta：”我只是用眼镜拍孩子，分享给祖父母”，被 sjamaan 直接打脸：
“You might want to think twice about this. The videos from those glasses are uploaded and used for AI training, for which they are also reviewed by humans in Kenya, according to the explainer from Bits Of Freedom.”
— sjamaan [c:49828314]
（译文：你再想想。这些眼镜拍下的视频会上传并用于 AI 训练，训练数据由 Kenya 的人工审核——这是荷兰 Bits Of Freedom 组织的解释视频里说的。）
GrinningFool 跟进，引用 BBC 报道，结论直接：
“Did you think they were making it up? I couldn’t conceive of a world in which I’d provide meta with video of my family. Or… anything, really.”
— GrinningFool [c:49829321]
（译文：你以为他们编的？我无法想象一个我会主动把家庭视频交给 Meta 的世界。任何东西都不会交。）
但关于罚款威慑力，HN 算了一笔账：ElProlactin 指出加州 CCPA 对 Facebook 的累计罚款至今是 $0：
“a quick search indicates that, to date, Facebook has paid $0 in fines to California for CCPA violations. The biggest fine ever recorded for CCPA violations was against Disney, and it was for a paltry $2.75 million. Disney pulls in $2.75 million in revenue every ~10 minutes.”
— ElProlactin [c:49828955]
（译文：快速查证，截至今日 Facebook 因违反 CCPA 向加州支付的罚款总额为 $0。史上最高的 CCPA 罚款是罚 Disney，275 万美元——Disney 每 10 分钟就赚这么多。）
piva00 把这一论点推向极致——罚款机制对 Meta 几乎无效：
“Meta pulls >100B USD in profits, an added fine of even a few hundreds millions USD could be worth it depending on cost-benefit. … initial fine and a requirement to change behaviour, no change in behaviour increases the fine up to 6% global revenue, and the fines will keep coming until the behaviour is changed.”
— piva00 [c:49829096]
（译文：Meta 年利润超过 1000 亿美元，几亿美元的罚款对它而言是成本收益可接受的……真正的牙齿应该是欧盟那套：初犯罚款 + 强制改变行为，不改则按全球营收 6% 累加，直到改了为止。）
平台不是公共基础设施——但它在管公共话语
pino83 把讨论拔高一层：Meta 删自己的内容，本来就是它作为私营公司的权利。
“But we are talking about Meta removing content from their own website, right? In contrast to what people think, Facebook IS NOT public infrastructure. If Meta removes some content from Meta website, and you don’t like that, it’s a you problem, right?”
— pino83 [c:49829047]
（译文：我们说的是 Meta 从它自己的网站删内容，对吧？跟很多人想的不一样，Facebook 不是公共基础设施。Meta 在自己的网站下架内容，你不喜欢，那是你的问题。）
fn-mote 直接指出这种”既要免责又要编辑权”的矛盾：
“They want protection from liability and also editorial control? Regardless of the legal status, it seems clear that when massive companies refuse to allow the media they own to publish a dissenting opinion, we’re not headed towards a
moredemocratic society.”— fn-mote [c:49829313]
（译文：它们既想要责任豁免，又想要编辑权？抛开法律状态不谈，大公司拒绝发布异议内容，社会明显不在往更民主的方向走。）
xeyownt 的立场最尖锐——逼 Meta 选一个：
“No problem with that as long as content responsibility is enlarged to all videos published on their platform. … They have to choose, safe harbor provisioning or responsibility for the content.”
— xeyownt [c:49829344]
（译文：可以，但前提是责任扩大到平台上所有视频……它们必须二选一：要么享受安全港豁免，要么为内容负全责。）
Streisand 效应：删了反而更红
tsak 一句话预测：
“Hopefully the Streisand effect will be in full force on this one.”
— tsak [c:49827980]
（译文：希望 Streisand 效应这次火力全开。）
而 the_gipsy 指出 Instagram 上”街头骚扰 prank”内容是 Meta 自身放出来的：
“Instagram is full of videos of guys harassing and antagonizing people in public as ‘pranks’. And this is not some coincidence, since both products are from the same company, they basically unleashed this for engagement.”
— the_gipsy [c:49828038]
（译文：Instagram 满屏都是男人在街头骚扰、挑衅别人，当作「整活」。这不是巧合——两款产品同一家公司，它们就是为了 engagement 把这玩意儿放出来的。）
asdff 的提议更实际：
“Great idea for protests outside meta offices. Just stand on the public right of way outside the entrance, smiling in your meta ray bans.”
— asdff [c:49828054]
（译文：在 Meta 办公楼外抗议的好主意——就站在门口的公共通行道上，戴着 Meta Ray-Bans 微笑。）
这算不算骚扰？——少数派的逆辩
少数派观点认为这件事的重点不是 Meta 下架，而是视频本身就是骚扰。yread 给出了 YouTube 链接并评论：
“He did harass them a bit. I’m looking forward to Meta taking down all videos where someone is a bit harassed. Internet will be a much nicer place!”
— yread [c:49828494]
（译文：他确实骚扰了他们一下。我很期待 Meta 下架所有「稍微骚扰」的视频，互联网会美好得多！）
rithdmc 看了视频后给出了更具体的反对：
“Having watched the video, this hardly meets the standard of ‘harassment’ or bullying. This is a vox pop, that people can (and do) walk away from.”
— rithdmc [c:49828143]
（译文：看完视频，这根本够不上「骚扰」或霸凌的标准。这是街头采访，人们可以（也确实会）走开。）
而 dezsiszabi 直接反对”骚扰”框架：
“I didn’t see any harassment. They made a product, interviewer is using it as intended. He walked up to people, asked questions while using their product. If they don’t like it, walk away. Zero harassment.”
— dezsiszabi [c:49829452]
（译文：没看到任何骚扰。他们造了这款产品，记者按设计用途使用——走上去提问。如果员工不喜欢，走开就是。零骚扰。）
CSAM 与海滩偷拍：最严重的”非骚扰”用例
讨论里反复出现一个事实——Meta 平台上有远比这恶劣的内容却不被处理。Mashimo 举了一个具体例子：
“I have read news article about people wearing meta glasses, removing the red recording LED and filming woman and girls at beaches in bikini without their knowledge. And these video often stay up on metas Instagram, accounts don’t get suspended. THAT is the issue.”
— Mashimo [c:49829957]
（译文：我看过新闻——有人戴 Meta 眼镜、拆掉红色录制指示灯，在海滩偷拍穿比基尼的女性和女孩。视频经常挂在 Meta 的 Instagram 上，账号也没被封。这才是问题。）
radicalbyte 走得更远，把”AI 训练”和”最坏滥用”串成一条线：
“It’s not ‘perfectly suited’, it’s EXPLICITLY DESIGNED TO do that. And it’s very well designed and very effective. But hey it’s OK because Meta only want to record everything to train their AI, who cares if some of their friends use it to create CSAM material, their third-world country workers upload it to PornHub and the NSA secretly pipe all of that through their servers.”
— radicalbyte [c:49829546]
（译文：不是「恰好适合」，它「被明确设计」做这件事。而且设计得非常好用。但没关系——Meta 只为训练 AI 录一切，谁在乎有些朋友的产物是 CSAM、第三世界审核员把内容上传 PornHub、NSA 把这些数据偷偷走它们的服务器。）
注：HN 用户 bennyelv 对此回复要求引用来源（”Citation needed buddy…” [c:49829880]），radicalbyte 未进一步举证。
典型观点一览
| 立场 | 用户 | 一句话 |
|---|---|---|
| 虚伪——Meta 容忍所有 Meta 眼镜拍的内容，唯独不能拍自己 | ElProlactin [c:49827967] | rules for thee, but not for me |
| 虚伪类比——卖车嫌车吵、做儿童产品不让自己孩子用 | suttontom [c:49829373] | “产品可以拍别人，拍我就不行” |
| 虚伪定名——爱泼斯坦眼镜 | tjpnz [c:49828511] | 给不在乎 consent 的人用 |
| 公开录像应合法——员工脸已模糊 | ElProlactin [c:49828097] | 肖像权诉求应由法院裁定，不该 Facebook |
| 公开录像各地差异巨大——荷兰/德国/爱尔兰规则不同 | lionkor [c:49828115] / rithdmc [c:49828151] | 欧盟内部没统一答案 |
| 隐私+AI——视频上传训练，Kenya 人工审核 | sjamaan [c:49828314] | “你再想想” |
| 罚款无效——Facebook 在加州 CCPA 累计罚款 $0 | ElProlactin [c:49828955] | 百万级罚款对它是零钱 |
| 平台不是公共基础设施——删内容是私营公司权利 | pino83 [c:49829047] | “不喜欢是你的问题” |
| 平台既想免责又想编辑权——选一个 | xeyownt [c:49829344] | 二选一 |
| Streisand 效应——删了反而更红 | tsak [c:49827980] | 一句话预测 |
| 视频构成骚扰——让员工被拍 | yread [c:49828494] | “期待 Meta 下架所有这种视频” |
| 反对骚扰框架——这是街头采访，不是霸凌 | rithdmc [c:49828143] | “人们可以走开” |
| 反对骚扰框架——产品按设计用途使用 | dezsiszabi [c:49829452] | “不喜欢就走开” |
| 真正的问题是海滩偷拍——拆掉 LED 拍比基尼视频不被处理 | Mashimo [c:49829957] | 平台选择性执法 |
| AI 眼镜为最坏滥用而设计——CSAM、NSA pipe | radicalbyte [c:49829546] | “被明确设计” |
总体情绪
整个 HN 讨论的核心反讽是：一家用户口中”年利润超过 1000 亿美元”的公司，对一个戴着自己产品拍自家员工的讽刺视频零容忍，对自己平台上拆 LED 海滩偷拍、AI 训练数据流到 Kenya 审核员、CSAM 衍生内容长期睁眼闭眼。
少数派（yread）承认视频带”轻微骚扰”色彩，”期待 Meta 下架所有这种视频”。rithdmc 和 dezsiszabi 反对这一框架——前者强调这是街头采访、人们可以走开；后者直接说”产品按设计用途使用、不喜欢就走开”。HN 多数派显然不接受”骚扰”叙事：他们认同的逻辑是”那你们应该停止制造让这种事成为可能的产品”。
法律层面的讨论揭示了一个尴尬现实：欧盟内部对”公共场所拍人”没统一答案。Meta 完全可以利用这一点在某些成员国卖、某些成员国禁——这种”逐国合规”策略本身又成了”不在乎 plebs 隐私”的证据。
删视频这件事本身或许只值 24 小时热度，但当一家公司选择性执法自己平台的内容、靠法律管辖差异在欧盟打擦边球——Streisand 效应只能解决单个视频，解决不了”删谁留谁全凭公司心情”的运作机制。
引用帖子
| # | 标题 | URL |
|---|---|---|
| 1 | Meta takes down a critical video about meta AI Glasses after filming at Meta | HN / Reddit |
本摘要中提及的言论均为 HN 评论用户的个人观点，不代表本站立场。
本摘要由 AI 模型辅助生成：minimax-cn-coding-plan/MiniMax-M3