视频简介
这个视频我将不从“教你用什么 AI 工具”开始，而是先带你退一步，真正看清 AI 到底是什么。在过去几个月里，AI 的变化速度非常快，很多人会产生一种焦虑：是不是不学 AI 就会被淘汰？是不是现在一定要转型？但在你做任何决定之前，你必须先理解 AI 的全貌，而不是只看到表层的 ChatGPT 或写代码工具。 在这期分享中，我会结合自己在银行、金融科技、IT 咨询和软件开发中的真实经验，从AI 的发展历史、技术分类、应用层级一步步拆解。你会看到，AI 并不是最近才出现的东西，从 DeepMind、AlphaGo，到语音助手、计算机视觉、RPA，再到今天的 LLM 和 AI Agent，其实是一条非常清晰的演进路线。 我也会重点讲一个大家很关心但很少被认真讨论的问题：AI 的经济价值和成本结构。除了大家熟悉的 token 成本，真正昂贵的是算力、能源、硬件折旧，以及背后的金融风险。这也是为什么近几个月会不断有人提到“AI 泡沫”，以及它和当年的 dot-com 泡沫有哪些相似、又有哪些本质不同。 最后，我希望你在看完这期视频后，不是得到一个“标准答案”，而是能回到自己：你的背景是什么？你的行业能用 AI 做什么？你现在用 AI，是在提高生产力，还是只是在跟风？ 这才是 1 月 22 日这次分享真正想留给大家的问题。 In this video, I will take a step back from tools and tutorials, and focus on something more fundamental: what AI really is, and how we should think about it before rushing into “AI transformation.” Over the past few months, AI has evolved extremely fast, creating both excitement and anxiety. Many people feel they must learn AI immediately or risk falling behind—but that mindset often leads to shallow understanding. Drawing from real experience in banking, fintech, IT consulting, and software development, I will walk through the broader AI landscape—from early systems like DeepMind and AlphaGo, to voice assistants, computer vision, RPA, large language models, and now AI agents acting as digital co-workers. AI is not one single technology; it is a layered ecosystem with very different applications and value propositions. We will also explore a topic that is often ignored in online discussions: the economic value and cost of AI. Beyond token pricing, real AI systems require massive compute power, energy consumption, specialized hardware (GPUs and TPUs), and complex financial structures. This is why some economists and investors are raising concerns about an “AI bubble,” and why comparisons to the dot-com era keep resurfacing. By the end of this video, my goal is not to tell you whether you should pivot to AI, but to help you build the judgment needed to answer that question yourself. Understanding where AI truly creates value—and where it does not—is far more important than blindly following trends.
