Nikkei 225 Index Falls Expanding to 3% 日经225指数跌幅扩大至3%

The departure of OpenAI's chip veteran to join Anthropic is far more than an ordinary personnel change. It is clearly a battle for the "foundation" as the AI arms race enters deeper waters. Chips are the hardware heart of AI, and the movement of top chip architects signals potential shifts in underlying technical routes or a collective reevaluation of existing technological paths. Anthropic has always emphasized its philosophy of "safe AI." Is this move a victory for technological idealism, or simply a poaching achieved with higher salaries and a more "correct" narrative? Regardless, this slap lands loudly on the faces of those giants who thought money could buy the future of AI. The soul of technology ultimately resides in people, and people will always choose where they can better realize their vision.

On the other side, the success secret of that supermarket nicknamed "Little Sam" forms a pungent contrast with the "high-end, grand" image of AI. It sells not computing power or models, but extremely cost-effective daily necessities and a streamlined SKU range. This scene is like a dull blow to the heads of all business models obsessed with using AI to "empower everything." When tech companies are passionately discussing in meeting rooms how large models will reshape retail, the most surging consumer power is voting with their feet, crowding into those brick-and-mortar warehouses with glaring lights and towering shelves that seem disconnected from the digital age. AI may optimize supply chains, but what truly pries open wallets is the naked, technology-threshold-free appeal of being "cheap and generous." The AI revolution, chased by capital, appears somewhat pale before the shelves of everyday life.

Most intriguing is that trillion-dollar AI company that "prohibits the use of AI during interviews." This is a perfect parable for our times. While selling the most powerful AI tools to the world and shaping a future where AI can do anything, it strictly forbids candidates from using these tools when selecting talent internally. This is no different from an arms dealer requiring soldiers to engage in hand-to-hand combat when recruiting them. It reveals a cold fact: at the current stage, humanity's core creativity, deep thinking ability, and unique resilience in problem-solving are still what these top AI companies most crave and most fear being replaced by AI. They sell AI but are more vigilant than anyone about its encroachment. This self-contradiction is precisely the most authentic portrayal of this era—we are both the creators of AI and might become its first unemployed sacrifices.

Returning to CITIC Securities' optimistic report on the insurance industry. The report is full of flashy terms like "risk-free allocation opportunities," "high certainty of profitability," and "quasi-utility barriers." But if we strip away this rhetoric and examine it in the context of the various paradoxes triggered by AI, we discover a profound disconnect. Financial capital is still trying to price the AI-driven "new economy" using the valuation logic of the old world, even imagining services like elderly care and medical care—those most in need of human touch—as "barriers" that can be built with AI. However, when an AI giant itself has to admit that human "primitive" abilities are irreplaceable in interviews, why should we believe that soulless algorithms can truly build trust barriers concerning matters of life and death?

Therefore, today's several news items piece together a divided map of the AI world: capital is reveling, technology is in internal strife, consumers are returning to primitive needs, and the most cutting-edge explorers are defending human values in the most traditional ways. The decline in the Nikkei Index may simply be one of the global market's instinctive reactions to this enormous uncertainty. AI will not bring simple poverty or wealth; it will tear the world apart even more completely—some will use AI to earn unimaginable wealth, others will fall into deeper confusion and poverty because of AI, and most people may ultimately choose, like the customers flocking to "Little Sam," to vote with their feet and walk toward those human realities that AI cannot yet simplify.

日经225指数今天又跌了3%，资本市场的风声鹤唳总能第一时间牵动全球神经。但比这更让人心里一沉的，是几条看似不相关的AI资讯拼凑出的诡异图景：万亿美元估值的OpenAI，其核心芯片元老却跑去了死对头Anthropic；一个名叫“小山姆”的仓储式超市在中国商场爆卖，背后的商业逻辑与AI毫不相干；而那篇刷屏的《用了AI之后，公司好像更穷了》文章，恐怕才真正戳破了无数企业主的幻想。

OpenAI芯片元老的离职并加入Anthropic，远不止是一次普通的人事变动。这分明是AI军备竞赛进入深水区后，对“根基”的争夺。芯片是AI的硬件心脏，顶级芯片架构师的流动，预示着底层技术路线可能发生动摇，或是对现有技术路径的集体反思。Anthropic一直强调其“安全AI”的理念，此举是技术理想主义的胜利，还是仅仅用更高的薪水和更“正确”的故事完成了挖角？无论如何，这记耳光响亮地打在了那些以为有钱就能买到AI未来的巨头脸上。技术的灵魂，终究是依附于人的，而人，永远会选择更能实现其愿景的地方。

另一边，那家被称为“小山姆”的超市，其成功秘诀与AI的“高大上”形成辛辣对比。它卖的不是算力，不是模型，而是极致性价比的日用百货和精简的SKU。这场景像一记闷棍，敲在所有痴迷于用AI“赋能一切”的商业模式头上。当科技公司在会议室里激情澎湃地讨论大模型如何重塑零售时，最汹涌的消费力量，正在用脚投票，挤进那些灯光刺眼、货架高耸、仿佛与数字时代脱节的实体仓库里。AI或许能优化供应链，但真正撬动钱包的，是那种赤裸裸的、毫无技术门槛的“便宜大碗”的吸引力。资本追捧的AI革命，在人间烟火的货架前，显得有点苍白。

而最值得玩味的，是那家“面试时禁止使用AI”的万亿美元AI公司。这简直是一个完美的时代寓言。它一边向全世界兜售最强大的AI工具，塑造一个AI无所不能的未来；一边在内部选拔人才时，严厉禁止候选人使用这些工具。这无异于一个军火商在招聘士兵时，要求他们必须徒手格斗。它揭露了一个冰冷的事实：在当前阶段，人类最核心的创造力、深度思考能力和解决问题的独特韧性，仍然是这些顶尖AI公司最为渴求、也最为恐惧被AI替代的东西。他们售卖AI，却比任何人都更警惕AI的侵蚀。这种自相矛盾，恰恰是这个时代最真实的写照——我们既是AI的造物主，也可能成为它第一批失业的祭品。

回到中信证券那份对保险业的乐观报告。报告里充满了“无风险配置机会”、“高盈利确定性”、“准公用事业壁垒”这类光鲜词汇。但如果我们剥离这些话术，将其放在上述AI引发的种种悖论中审视，会发现一种深刻的割裂。金融资本仍在用旧世界的估值逻辑，试图为AI驱动的“新经济”定价，甚至将养老、医疗这些最需要人情味的服务，也想象成可以靠AI构筑的“壁垒”。可是，当一家AI巨头自己都不得不承认，面试中人类的“原始”能力不可替代时，我们凭什么相信，那些缺乏灵魂的算法，能真正构筑起关乎生老病死的信任壁垒？

所以，今天的几条资讯，拼凑出了一幅分裂的AI世界地图：资本在狂欢，技术在内讧，消费者在回归原始需求，而最前沿的探索者，正在用最传统的方式捍卫人性的价值。日经指数的下跌，或许只是全球市场对这种巨大不确定性的本能反应之一。AI不会带来简单的穷或富，它会将世界撕裂得更加彻底——一部分人用AI赚取难以想象的财富，另一部分人因AI陷入更深的困惑与贫困，而大多数人，可能像涌向“小山姆”的顾客一样，最终选择用脚投票，走向那些AI暂时还无法简化的人间真实。