Terry
Private Fund Portfolio Manager & Trader
Tencent is an underrated way to play AI agents. Meta's Muse, an agent that sends emails, compares prices and books flights, topped the US App Store with 2.5M downloads in two weeks. WeChat's Xiaowei follows almost the same product logic, but WeChat has a social graph, mini-programs and payments, a moat Meta lacks. The business model is unproven: Muse's free tier allows 100M tokens a week, heavy users cost $20-40, and closing the commercial loop may be harder than the technology. x.comÂ
Tencent is priced for more pessimism than is warranted, making it a good bet. Heavy Q1 deferred revenue should be released mostly in Q2 and smooth full-year results, pushing earnings pressure out to Q4, so the coming period should trade on Tencent's AI. Its WorkBuddy product keeps it out of the AI losers' group, and among very large consumer companies its reported results look strong without relying on channel-inventory swings to pass as growth. x.comÂ
Meta's established consumer distribution could give Muse an advantage in bringing AI agents to everyday users. The stock looks increasingly constructive heading into tomorrow's earnings. Muse has 10 million downloads, while Meta's apps reach billions and its ad network connects millions of businesses. If agents become a consumer interface for discovery and transactions, the economics could extend into commerce, advertising and business services. x.comÂ
Tencent owns one of the best closed loops from intent to identity to payment to fulfillment, yet trades as if AI agents will bypass it. Weixin combines 1.44 billion users with chat, payments, Mini Programs and merchant reach, so its Xiaowei agent can hand work to services that already know the user, a distribution and return advantage. About $75 billion of cash and $128 billion of investments against a $491 billion market cap leave the operating business near 7x roughly $40 billion of earnings. … Show more
Meta's Muse is likely to gain a major consumer-agent advantage from a network linking its users' agents. Meta's 3.5bn daily platform users offer unmatched distribution; 25% of 10mn Muse signups use it daily, with early retention twice Threads' initial rate. Muse is the best pure consumer agent today, and the odds of a rival surpassing it are low, though it is early and execution matters. The stock prices in little beyond Meta's core business. x.comÂ
Meta is having its iPhone moment with Muse, while the market still sees a social media company overspending on AI. Muse is a personal assistant that can shop, book travel, schedule appointments and send emails, through its own app or inside WhatsApp, with Ray-Ban glasses next. Meta is one of the only companies with all the pieces, and Muse can sit on top of Facebook, Instagram, WhatsApp, Threads and the glasses. The market fixated on capex and missed what the money was building. x.comÂ
Meta's focus on personal and knowledge work makes its $5T consumer-AI path easier to see because those are the tasks its agent must handle. Its company-designed index scores Wonder Full at 59.1 versus 55.2 for the next best model shown, though such benchmarks warrant caution. Ads can subsidize Muse usage across Meta's apps and glasses. Monetizing intent depends on retention and real-world task completion; computer use, healthcare and some coding remain weak. x.comÂ
Meta's hundreds of billions in AI investment have produced Muse, a free, largely undifferentiated personal assistant app that will likely never generate economics. Muse topped app-store downloads and helped fuel a two-day gamma squeeze mostly because it is new and free. Amazon has already cut it off from shopping its site, and once the novelty fades it will probably see little use given its limited utility; people actually like doing their own shopping. x.comÂ
Meta's Muse looks like the new ChatGPT moment, reportedly outpacing ChatGPT's mobile launch on several metrics, and Meta's stock should run to a bigger percentage gain. x.com x.comÂ
Meta's billions of users, distribution and integrated ecosystem could let its increasingly capable AI models reach extraordinary scale. Muse Wonder and Muse Wonder Full bring memory, focus and task execution, though Meta's reported model results still need independent validation. Meta is one of the highest-conviction stocks: as a megacap its upside may be less explosive than smaller names, but its risk/reward looks more favorable with potentially less downside in this environment. x.comÂ
Meta's earnings offer little near-term upside because Muse user growth is mostly priced in and monetization remains vague. The market wants AI compute to turn into profit. Retention or paid-share commentary could help, but Zuckerberg is unlikely to provide it; SpaceX says Grokbot has 800K paid users, a high bar. Meta remains attractive long term. Higher AI capex guidance would show confidence Muse can repay the spending and be bullish for the whole AI infrastructure stack. x.comÂ
Meta's Muse can bring AI to the masses by making it easy for everyday people, including non-technical users and those who do not yet trust AI, rather than by topping every benchmark. Meta already reaches 3.6 billion people a day across its apps, an enormous distribution advantage, and paired with a great user experience and reliable results, Muse could become a huge business. x.comÂ
