Meta庞大的用户基础和AI增强的广告业务支撑其长期增长潜力，尽管高额资本开支和营业利润率下滑带来风险。
逐段跳播
今天，我将谈谈 AMD、Broadcom、Amazon、Meta 和 Nvidia。Amazon、Microsoft、Google 和 Meta 正花费数十亿美元建设数据中心、购买芯片并确保电力供应。
Today , I will talk about AMD, Broadcom, Amazon, Meta, and Nvidia. Amazon, Microsoft, Google and Meta are spending billions to build data centers, buy chips and secure electricity.
我们最近可以看到 Meta 的股价涨得有多高，而 Meta 主要依赖广告收入。
We can see recently how high Meta's stock has risen, and Meta relies mainly on advertising revenue.
现在让我们进入下一只股票，“Meta”。与我在这档节目里反复谈到的半导体公司相比，Meta 在人工智能方面是一种完全不同的投资。它不需要向你出售图形处理器，也不需要向你出租数据中心，就能从人工智能中获益。它已经拥有了一样极其宝贵的东西：用户的注意力。
Now let's move on to our next stock, "Meta". Meta is a completely different investment in artificial intelligence than the semiconductor companies I've talked about so much on my channel.
It doesn't need to sell you a graphics processing unit or rent you a data center to benefit from artificial intelligence. It already possesses something incredibly valuable: the attention of users.
Facebook、Instagram、WhatsApp 和 Messenger 加起来每天触达近 36 亿人。
Facebook, Instagram, WhatsApp and Messenger together reach nearly 3.6 billion people every day.
当你拥有这样的触达规模时，即便盈利能力只有相对较小的提升，也能带来数十亿美元的额外收入。我从争夺注意力的战争这一角度来看待 Meta。每家公司都想获得更多客户，而这些客户把很大一部分时间花在社交媒体上。
When you have this kind of reach, even relatively small improvements in profitability can generate billions of dollars in additional revenue. I think of Meta from the perspective of the attention-grabbing war.
Every company wants to get more customers, and these customers spend a large part of their time on social media.
如你所见，“Bloom”不是“Bloom Energy”，而是一家饮料公司，你知道的。它就像一种气泡饮料。我不喝它，但我看到他们有 5 亿美元的收入，而且投放了大量广告。
As you saw, "Bloom" is not "Bloom Energy," but a beverage company, as you know. It's like a fizzy drink . I don't drink it, but I saw that they have $500 million in revenue and they run a lot of advertisements.
因此，每一家真正想要增长的公司，尤其是大公司，都依赖 Meta 的广告。Meta 利用人工智能提升其广告产品的效率。它的算法可以推荐更相关的内容、识别潜在客户，并帮助广告主改进其广告活动。
Therefore, every company that truly wants to grow, especially large companies, relies on meta ads. Meta uses artificial intelligence to improve the efficiency of its advertising products.
Its algorithms can suggest more relevant content, identify potential customers, and help advertisers improve their campaigns.
数据已经表明，这项工作对 Meta 来说有多么强大。自然，所有在 Meta 上从广告中获益的公司都会继续投放广告，甚至会扩大投放。我所说的扩大，是指它们未来会在广告上花更多钱。
The numbers already show just how powerful this work is for Meta. Naturally, all companies on Meta that benefit from advertising will continue to advertise, and even expand upon it.
What I mean by expansion is that they will spend more on advertising in the future.
那么，让我们来看看 Meta 的业务。在 2026 年第二季度，Meta 实现了 608 亿美元的收入，同比增长 28%。
So, let's take a look at Meta's business. In the second quarter of 2026, Meta generated revenues of $60.8 billion, a 28 % year-over-year increase.
广告展示量增长了 14%，而平均广告价格上涨了 12%。字面地想一想这件事。Meta 不仅展示了更多广告，还从每条广告中获得了更多收入。当这两件事同时发生时，你就能实现显著的收入增长，而实际上并不需要以同样的速度扩大用户基础。
Ad impressions increased by 14%, while the average ad price increased by 12%. Think about this literally. Meta not only displays more ads, but also generates more revenue from each ad .
When these two things happen together, you can achieve significant revenue growth without really needing to increase your user base at the same rate.
现在，Meta 也在尝试建立一项超越广告的业务。其最新的 AI 产品，包括新闻，在个人助理、商业自动化和购物领域提供了机会。
Now, Meta is also trying to build a business that goes beyond advertising. Its latest AI products, including news, offer opportunities in the areas of personal assistance, business automation, and shopping.
这就是这件事为什么有意思。想象一下，一位消费者让智能助手找到某件特定产品、比较不同选项并完成购买。Meta 已经拥有数十亿使用其应用的潜在客户。如果它能成功地将购物和其他交易整合进其 AI 产品，它就可能开拓出超越传统广告的新收入来源。
Here's why this is interesting. Imagine a consumer asking a smart assistant to find a specific product, compare different options, and complete the purchase. Meta already has billions of potential customers who use its apps.
If it can successfully integrate shopping and other transactions into its AI products , it could develop new revenue streams that go beyond traditional advertising.
但现在还为时过早。数百万人试用过一款 AI 产品，这一事实并不自动意味着这些人会成为付费客户。Meta 也面临着激烈的竞争。
But it's still too early . The fact that millions of people have tried an AI product does not automatically mean that these people will become paying customers. Meta also faces significant competition.
现在，还有另一个重大风险。Meta 预计 2026 年的资本支出将在 1300 亿至 1450 亿美元之间。这是一笔非常庞大的金额。
Now, there is another major risk as well. Meta expects spending between 130 billion and 145 billion on capital expenditures in 2026. That is a very large amount of money.
第二季度，营业利润率从去年同期的 43% 降至 31%，反映出费用上升。所以，问题不在于 Meta 是否有能力创造收入，而在于所有额外的 AI 支出能否带来足够的增量收入和营业利润，以证明这笔投资是合理的。
In the second quarter, the operating margin fell to 31% from 43% a year earlier, reflecting higher expenses. So, the question is not whether Meta is capable of generating revenue, but whether all the additional AI spending will produce sufficient incremental revenue and operating profit to justify the investment.
我会关注广告业务的增长。AI 产品的采用情况和营业利润率也值得密切关注，但业务的主要驱动力当然是广告。
I will monitor the growth of the advertisements. The adoption of AI products and operating margins are also important to monitor closely, but the main driver of business is, of course, advertising.
我的五只股票：AMD、Broadcom、Amazon、Meta 和 Nvidia。这五家公司都能以不同方式利用人工智能，我相信它们具有巨大的长期潜力。当出现有吸引力的机会时，我会继续买入。
My five stocks: AMD, Broadcom, Amazon, Meta, and Nvidia. All five companies are in a position to leverage artificial intelligence in different ways, and I believe they have great long-term potential.
I will continue to buy when attractive opportunities arise .
观察点
这个频道对 $META 的历次判断
Invest with Henry 在这只股票上共 3 条判断，这里只列相邻的几条。