Meta (META) Q3 2026 Earnings Breakdown: Muse AI Agent, $64.6B Revenue Beat & Class C Stock Split
Publicado no IA Dots ·
Agentes de IAAI Signaldot-ai_agentsdot-artificial_intelligencedot-metadot-museidioma-enInglêsMuseyt-ZNXaV0rbcHo
Vídeo em Inglês, publicado pelo canal Big Picture Investing With Kevin McCormick.
About this content: The summary, key points and topics below are generated automatically from the video transcript. They reflect the content, opinions, claims and predictions presented in the video and do not represent a position, validation or fact-check by IA Dots.
Video summary
Meta's recent earnings report highlights strong revenue growth and the rapid adoption of its new AI agent, Muse, alongside significant capital expenditures, legal charges, and a drop in free cash flow.
Key points
- Meta launched Muse, a personal AI agent that reached 10 million users in the US and Canada within six weeks, with a 25% daily usage rate.
- Total revenue reached $64.6 billion, a 26% increase year-over-year, driven by a 10% growth in ad impressions and a 14% rise in average ad prices.
- GAAP operating income fell 37% to $13 billion, impacted by a $9.9 billion legal charge related to a settlement with state attorneys general.
- Free cash flow turned negative at $2.1 billion for the quarter, while capital spending and quarterly expenses surged due to heavy investments in computing and AI.
- Meta announced a new non-voting class C stock, functioning as a two-for-one split to facilitate future capital raising, compensation, or acquisitions.
Topics
Meta EarningsMuse AI AgentRevenue and Ad GrowthCapital ExpenditureFree Cash FlowLegal SettlementStock Split
Video transcript
Transcript from available captions; it may contain errors.
Welcome back to Big Picture Investing. I'm your host, Kevin McCormick. 10 million people, six weeks, and a quarter of them are using it every single day. That's Muse, Meta's new personal AI agent, and it's the reason everyone is talking about this earnings report. But here's the twist. In the same quarter, Meta launched its most successful product in years. GAP operating income fell 37% and free cash flow went negative. So which story is the real one? Here's the bottom line up front. Meta, ticker MA, beat on revenue, guided the fourth quarter above the street, and the stock was up roughly 3 to 5% after hours in early trading. Underneath that, you've got a $9.9 billion legal charge. capital spending of almost $ 35 billion in one quarter and a brand new non- voting share class that works like a twofor one split. Here's the road map. First muse because that's the story. Then the numbers, then guidance and the cash burn. Then what management said on the call, then the bull versus bear debate, three full rounds, and finally the big picture. Let's start with Muse. On the call, Mark Zuckerberg said, quote, "In just a month and a half, more than 10 million people in the US and Canada have set up Muse Personal Agents, and that's only two countries. Meta hasn't launched it globally yet, partly because it's so expensive to run. The number I'd focus on isn't the 10 million, though, it's the 25%." Zuckerberg said about a quarter of people who set up Muse use it on any given day. and he said that's more than twice the level thread started at. Downloads are easy. Daily habits are hard. That's the signal that this might be a real product and not a novelty. And what are people doing with it? Shopping. More than a quarter of Muse users are shopping through it with an average purchase size of over $200. Zuckerberg said that in the first week, Meta had to work with Stripe to raise credit limits because the volumes were above what anyone expected. Here's why that matters to investors. An AI agent that people chat with is a cost center. An AI agent that people buy things through is a business. So, how does Meta get paid? Zuckerberg laid out the model. For most people, Muse stays free and Meta takes a fee on the transactions it helps facilitate, paid by businesses. Power users who want the most advanced models or extra tokens pay a subscription. Businesses pay for business features. And he went further. He said Meta's family of apps already earns several hundred a year per American user, and he expects Muse to reach an even higher level over time. That's a big claim and it comes with a big catch. Muse is in Zuckerberg's words very compute inensive to run. The plan is to improve its cost efficiency by more than 10 times in the next year through off- peak capacity, smaller models, and inference optimization. If they hit that, Muse can be profitable at scale. If they don't, every new user costs money. And that brings us to the numbers because the cost side of this story is already showing up. Total revenue was 64.6 billion, up 26% from a year ago. Wall Street was looking for about 63.4 billion. So that's a beat of roughly 1.2 billion. Meta's own guide was 61 to 64 billion. So it cleared the top of its range. What drove it? Ad impressions grew 10% and the average price per ad grew 14%. That price number is the important one. It means advertisers are paying more for each ad because Meta's AI targeting keeps getting better at turning ads into sales. The core engine isn't slowing down to pay for Muse, it's funding Muse. Now, the ugly part. Total expenses jumped 68% to 51.6 billion. That includes a $9.89 billion charge tied to Meta's August settlement with 52 state attorneys general over teen mental health. So, GAAP operating income fell 37% to about $13 billion and operating margin dropped to 20% from 40% a year ago. GAAP EPS was $3.83. Strip out the legal charge and Meta says EPS would have been $7.40. 40s. Consensus numbers are all over the place tonight. Roughly 550 to 650 depending on the source. So on an underlying basis, this looks like a solid beat. But the charge is real cash. The settlement provides for payments of up to about $18.6 billion in total. Which brings us to guidance and why the stock still went up. For the fourth quarter, Metaguided revenue of 73 to76 billion. The street was at about 73.75 billion. So the midpoint 74 a.5 is about 1% above expectations and it implies roughly 24% growth. For a company this size, that's remarkable. On spending, Meta raised its fullear expense outlook to 175 to 179 billion, mostly to absorb the legal charge and capex for 2026 was narrowed to 135 to 145 billion, up from a low end of 130 billion. Now, zoom out to cash because this is where the bare case lives. Capital spending in the quarter including finance leases was $34.9 billion. Operating cash flow was $ 32.8 billion. So free cash flow was negative $2.1 billion. A year ago it was positive 10.6 billion. Meta did not buy back a single share this quarter. And through 9 months they've spent about 86 billion on capex which means the fourth quarter alone could be 50 billion or more. Meta still has 82.5 billion in cash and securities against about 81 billion of long-term debt. That's not a crisis. But the cash machine that funded buybacks for a decade is now fully redirected into compute. And that's exactly why the new share class matters. Meta is creating non- voting class C stock distributed one for one to every existing shareholder. effectively a twofor one split with trading expected to start in late January. CFO Susan Lee said the purpose is a stock usable for capital raising, employee compensation acquisitions while preserving existing voting power. Translation: Meta is building a funding tool before it needs one. No financing plans yet, but they built the door. One more piece of good news before the debate. Reality Labs lost $4.1 billion this quarter, but Meta now expects losses of 18 to 19 billion this year, falling to around 15 billion next year as money shifts toward AI. Muse is coming to Metaglasses starting next month. So, with all of that on the table, it's time for the bull versus bear debate. Round one, is Muse a real business? Bulls open with the adoption data. 10 million setups in six weeks, 25% daily use, and $200 average purchases. Zuckerberg said Plaid told Meta it hadn't seen activity like this on any other service. If Meta takes even a small cut of that commerce at billions of users, it's a new revenue line on top of ads. Bear's answer. Meta gave no cost numbers. When Barclays asked about the per user cost of Muse, Susan Lee said, "We don't have specific numbers to share. A product that loses money on every user can grow fast and still destroy value. The whole thesis rests on that 10 times efficiency gain. Bulls come back." Zuckerberg said a year from now he expects the same intelligence to run at a fraction of the cost in a smaller model. And Meta already cut content processing costs by more than 80% in offline tests of its own Muse models. Round two, the cash bears say, "Look at the cash flow statement. Negative 2.1 billion in free cash flow, zero buybacks, capex guided as high as 145 billion, and management says it will keep investing aggressively in 2027, plus a new share class built for raising money. That's dilution risk. Bulls respond. Revenue grew 26% and the fourth quarter guide is above the street. Excluding the legal charge, operating income would have grown 11%. This is a company spending from strength, not desperation. Round three, legal risk. Bears point to the $9.9 billion charge, settlement payments of up to about 18.6 6 billion and Meta's own warning about youth related trials scheduled in the coming months. Bulls answer that this charge is largely one-time and the core ad engine just priced ads 14% higher. So what settles it? Three things. First, Muse's global rollout and any first monetization numbers. Second, the 2027 capex number on the January call. And third, whether Meta actually sells any class C stock. Zooming out to the big picture, every big tech company is spending enormous sums on AI. The difference is that Meta just showed a consumer AI agent with real daily usage and real commerce attached. That's what investors have been waiting for. Proof that the spending turns into a product people pay through. The risk is that the bill shows up long before the revenue does. Negative free cash flow, a new share class, and legal overhang are the cost of that bet. So, here's my question for you. Is Muse the next billion user platform or the most expensive experiment in tech? Let me know in the comments. And if this breakdown helped, hit subscribe so you don't miss the next one. Thanks for watching Big Picture Investing. I'll see you in the next one.
Assistir ao vídeo original no YouTube
Assuntos: Agentes de IA, Muse. Publicado no YouTube em 2026-10-21.
Seleção automática por assunto e idioma. O vídeo pertence ao canal de origem.
More from Meta's official Q3 2026 release
Meta Platforms published its results for the quarter ended September 30, 2026 on October 21, 2026. In the official release, Mark Zuckerberg highlighted the launch of Muse and of Muse Wonder, described as Meta's first frontier model. Net income was $9.86 billion and diluted earnings per share were $3.83, compared with $2.71 billion and $1.05 a year earlier. The $9.89 billion charge comes from an agreement announced in August 2026 with 52 attorneys general to resolve litigation related to teen mental health and social media, which provides for payments of up to about $18.59 billion; excluding that charge, Meta says income from operations would have been $22.88 billion and diluted EPS $7.40. Family daily active people averaged 3.64 billion in September 2026 (up 3%), capital expenditures were $34.91 billion, cash and marketable securities reached $82.45 billion, and headcount was 66,775, down 15% year over year.
For the fourth quarter of 2026, Meta expects revenue of $73–76 billion, and it now expects full-year 2026 total expenses of $175–179 billion, capital expenditures of $135–145 billion and Reality Labs losses of $18–19 billion. On the new non-voting Class C common stock, the release says Meta plans an effective date of January 19, 2027, followed by a one-time dividend of one Class C share for each outstanding Class A and Class B share, with record and payment dates to be set later by the Board.
Fact sheet
- Company: Meta Platforms, Inc. (Nasdaq: META)
- Release date: October 21, 2026
- Net income / diluted EPS: $9.86 billion / $3.83
- Excluding legal charge: Operating income $22.88 billion; EPS $7.40
- Daily active people: 3.64 billion (September 2026)
- Headcount: 66,775 (down 15%)
- Q4 2026 revenue outlook: $73–76 billion
- Class C stock: Effective date planned for January 19, 2027
Read also
- How Meta got ahead of OpenAI on AI agents | Decoder
- Mark Zuckerberg on how Meta's Muse AI agent will make people money
- How Meta's New AI Agent Is Different
Sources and references: Meta Investor Relations – Meta Reports Third Quarter 2026 Results · Wikidata – Meta Platforms (Q380) · video by the Big Picture Investing With Kevin McCormick channel on YouTube.