我的征尘是星辰大海。。。

The dirt and dust from my pilgrimage forms oceans of stars...

-------当记忆的篇章变得零碎，当追忆的图片变得模糊，我们只能求助于数字存储的永恒的回忆

作者:黄教授

手机视频列表

硅谷七姐妹分析之脸书

视频

音频

原始脚本

Meta 原 Facebook，AI 时代的社交场控权保卫战。
Meta 的核心焦虑本质是社交场域的主体和规则正在被 AI 重构。
当社交的参与者从真人向真人加数字人加 AI 账号转变，甚至未来可能完全由 AI 主导交互时，他担心自己会从社交规则制定者沦为被新规则淘汰的旧平台。
核心焦虑，社交主体变了，场的价值会丢吗？当下社交平台的死账号、虚拟账号、一人多号只是前奏，真正的威胁是，若未来70%以上的社交交互由数字人完成。
比如企业用 AI 账号做品牌运营，个人用数字人代聊、代维护关系，事务性沟通直接由 AI 对接 AI ，Meta 的社交场会失去核心价值吗？他的底气在于，只要有交互需求，就需要承载交互的场。
就像线下市场从真人摆摊到无人货架，只要买卖还在，市场的场地价值就还在。
但他的恐惧在于，若 AI 公司绕过他直接搭建 AI to AI 交互协议。
比如你让 AI 直接对接 客户的 AI 谈合作，全程不用登录任何社交平台，他的厂会被彻底架空。
这也是他早年压住元宇宙，如今死磕 AI 大模型的核心逻辑。
必须用技术掌控 AI 社交的规则，比如数字人创建标准、AI 交互安全机制，才能确保自己仍是厂的拥有 应对路径，用技术绑定守住厂的控制权。
一、用大模型做数字人管家，Meta 的 llama 系列开源模型，本质是降低数字人创建门槛，让更多企业和个人在他的平台，Facebook、Instagram、元宇宙 Horizon，利用他的模型做数字人。
相当于把 AI 社交的水源握在自己手里，避免用户跑到其他平台建数字人，分流交互场景。
二、用元宇宙锚定下一代社交形态。
从文字到图片、视频、直播，社交始终在向更贴近真人的实时交互进化。
元宇宙加 AI 是下一个形态，比如用 AI 驱动的数字人在元宇宙里开会、逛街。
Meta 他砸钱做元宇宙硬件，Quest 头显和 AI 交互，就是想提前站住下一代社交场，避免像当年错过短视频一样，被新形态的 AI 社交平台取代。
三、用 AI 加固广告基本盘，哪怕社交主体是数字人，企业仍需要通过平台触达背后的真人，比如数字人运营的品牌账号，最终还是 要吸引真人消费。
Meta 正在用 AI 优化广告精准度，比如识别数字人账号背后的真人需求。
同时探索数字人广告新模式，比如让 AI 生成的虚拟 KOL 在平台带货，确保核心的广告收入不被 AI 冲击。

修正脚本

英文翻译

Meta, formerly Facebook, is fighting a defensive battle for social platform control in the AI era.  
Meta's core anxiety essentially stems from the fact that the subjects and rules of the social domain are being reshaped by AI.

When social participants shift from real humans to a mix of real humans, digital humans, and AI accounts—or even when interactions in the future are entirely dominated by AI—Meta fears it will be downgraded from a rule-maker of social interactions to an outdated platform eliminated by new rules.

Its core anxiety: if the subjects of social interaction change, will the value of the platform be lost? The current dead accounts, virtual accounts, and multiple accounts per person on social platforms are just a prelude. The real threat is that if over 70% of social interactions in the future are carried out by digital humans—for example, companies using AI accounts for brand operations, individuals using digital humans to chat and maintain relationships on their behalf, or transactional communication directly between AI and AI—will Meta's social platform lose its core value?

Its confidence lies in the fact that as long as there is demand for interaction, there will be a need for a platform to host it—just like how offline markets evolved from human-run stalls to unmanned shelves; as long as buying and selling exist, the value of the marketplace remains.

But its fear is that if AI companies bypass Meta entirely and establish direct AI-to-AI interaction protocols—for example, letting your AI directly negotiate with your client's AI without ever logging into any social platform—its platform could be completely hollowed out.

This is the core logic behind why Meta bet on the metaverse early on and is now going all-in on AI large models. It must use technology to control the rules of AI social interaction—such as standards for creating digital humans and security mechanisms for AI interactions—to ensure it remains the owner of the platform.

**Response path: Use technology to lock in control of the platform.**

1. **Use large models as digital human butlers:** Meta's Llama series of open-source models are essentially designed to lower the barrier to creating digital humans, allowing more businesses and individuals to build digital humans on its platforms—Facebook, Instagram, and the metaverse Horizon—using its models. This is like holding the water source of AI social interaction in its own hands, preventing users from migrating to other platforms to create digital humans and diverting interaction scenarios.

2. **Anchor the next generation of social interaction with the metaverse:** From text to images, video, and live streaming, social interaction has always been evolving toward real-time, more human-like communication. The metaverse plus AI is the next form—for example, using AI-driven digital humans to hold meetings or go shopping in the metaverse. Meta is spending heavily on metaverse hardware (Quest headsets) and AI interaction to secure its position in the next-generation social platform, avoiding being replaced by new AI-powered social platforms as it was with short videos.

3. **Reinforce the advertising foundation with AI:** Even if the subjects of social interaction are digital humans, businesses still need to reach the real humans behind them through the platform. For example, brand accounts run by digital humans still need to attract real human consumption. Meta is using AI to optimize ad targeting—such as identifying the real human needs behind digital human accounts—while also exploring new digital human advertising models, like having AI-generated virtual KOLs promote products on the platform, ensuring that its core advertising revenue is not disrupted by AI.