Meta posts record Q3 revenue of $64.6 billion as AI investments drive engagement and commerce, though margins face pressure from legal and infrastructure costs.
Meta Platforms posted $64.6 billion in revenue for the third quarter of 2026, a 26.1 percent increase from the same period last year, according to its October 22 earnings release. Daily active people reached 3.64 billion, adding 100 million users year over year, while the operating margin fell to 20.1 percent from 40.1 percent in 2025. The company pointed to its Muse personal agent as a key engagement driver, noting that roughly a quarter of users who set it up interact with it daily. Read the full financial breakdown at finance.yahoo.com.
OpenAI took a contrasting approach at its September 29 DevDay, unveiling Dots as a paid agent for ChatGPT subscribers while Meta offers Muse at no cost. The Verge reported that Dots will initially reach only users on premium tiers, leaving most of ChatGPT’s 1.2 billion weekly users without access. Meanwhile Anthropic launched Claude Duet on October 21 as a research preview for Pro and Max customers in the United States, promising continuous background monitoring across more than 6,000 connectors. See OpenAI’s announcements at theverge.com.
The deeper story is how quarterly results are now measured by agent adoption rather than ad impressions alone. Meta’s free Muse strategy aims to lock in daily habits across shopping and productivity, while OpenAI’s subscription wall and Anthropic’s enterprise preview signal competing visions for who controls always-on workflows. At the same time, fresh funding for agent payment infrastructure and wearable devices suggests the ecosystem is moving faster than the regulatory frameworks meant to oversee it. This article examines whether engagement gains can offset margin pressure as always-on AI becomes a balance-sheet item.
The Rise of AI Agents and User Engagement
Meta's third-quarter revenue reached $64.6 billion, a 26.1% year-over-year increase, as its Muse personal agent reached roughly 25% of daily active users and helped drive commerce and productivity use cases finance.yahoo.com. Daily active people across the company's apps hit 3.64 billion, up 100 million from a year earlier, with management citing AI-powered features as a primary driver of higher engagement and video consumption on Facebook and Instagram. CEO Mark Zuckerberg noted that about a quarter of people who have set up Muse use it on any given day, signaling strong early retention for the free agent product.
While Meta offers Muse at no cost with a dedicated virtual machine for each user, rival OpenAI has taken a different approach by restricting its competing Dots agent to ChatGPT Pro subscribers paying $100 per month theverge.com. The Verge reported that Dots, powered by GPT-6 Astra, adopts a similar "disarming cuteness" design philosophy with customizable blob avatars but lacks the frictionless access that has fueled Muse's rapid adoption. Both companies are simultaneously betting on physical hardware devices to extend their agents beyond phones and computers, a category that has largely failed to gain traction so far.
The divergence in distribution strategy highlights a broader industry tension: whether AI agents become ubiquitous utilities or premium differentiators. Meta's ad-supported model aligns with free agent access that deepens engagement and improves recommendation quality, which CFO Susan Li confirmed is already delivering incremental advertising revenue. OpenAI's subscription gating reflects its revenue architecture but risks ceding mindshare in a market where network effects and data flywheels compound quickly. The hardware push from both firms suggests the next battleground will be form factor as much as capability.
OpenAI and Anthropic Compete on Agent Capabilities
OpenAI unveiled Dots at its September 29 DevDay conference in San Francisco, positioning the agent as a direct rival to Meta's Muse though initially limiting access to ChatGPT Pro, Business Premium, and Enterprise subscribers theverge.com. CEO Sam Altman demonstrated Dots building metaverse-like virtual worlds and emphasized its "real-deal AI" credentials powered by GPT-6 Astra, while acknowledging the product's cute, customizable blob interface mirrors Muse's approachability. The company also introduced a new $500-per-month ChatGPT Pro tier, signaling a tiered strategy for agent access.
Anthropic countered on October 21 with Claude Duet, a research preview for Pro and Max users in the United States that runs continuously to monitor notifications and connected apps while executing tasks in the background note.com. The system leverages more than 6,000 connectors to span email, accounting, sales, recruitment, and e-commerce operations, enabling cross-business workflows that go beyond single-app automation. Anthropic frames Duet as a shift from chat-based prompting to always-on agents that observe work status and act on established conditions, though the company cautions it is not yet ready for broad enterprise deployment.
Both firms are wagering that personality-driven software agents , cute, approachable, and always-on , will serve as the wedge for dedicated AI hardware, a category littered with false starts. OpenAI's hardware ambitions remain vague but were teased at DevDay alongside Dots' virtual world-building, while Anthropic's connector-heavy approach targets the enterprise process layer where switching costs are high. The competition is no longer just about model intelligence but about distribution, form factor, and the depth of integration into daily workflows, whether consumer or commercial.
Funding Winter and the AI Infrastructure Buildout
DeepSeek reportedly secured around $15 billion in funding on October 22, 2026, with Tencent Holdings and Contemporary Amperex Technology Co. Ltd. (CATL) among the participants, according to analyticsinsight.net. The massive capital infusion is intended to accelerate the development of large‑scale language models and data‑center infrastructure. Industry analysts view the round as a signal that even a tightening credit environment cannot stall core AI ambitions. The timing aligns with a broader trend of strategic bets on compute resources and model scaling. Such moves suggest that deep‑pocketed backers remain willing to underwrite risky, long‑term AI projects despite macro‑economic headwinds.
Nirva, an AI‑wearable startup, completed a $25 million Pre‑A round on October 22, 2026, bringing its total funding to $33 million, as reported by kucoin.com. The financing, led by undisclosed investors, follows a $15 million Series A for Mural that targets payment infrastructure for AI agents. Together, these rounds illustrate a fragmented but active capital market where niche hardware and platform layers attract distinct investor appetites. While DeepSeek’s megafund aims at foundational models, smaller ventures focus on specialized AI services and transaction layers. The diversification of funding targets reflects an emerging “stack‑level” investment strategy in the AI economy.
The juxtaposition of billion‑dollar megarounds and modest seed‑stage raises highlights a “funding winter” that is not uniformly cold; rather, it is selective, favoring projects with clear pathways to monetize AI‑driven infrastructure. Companies building the pipes,payment rails, edge devices, and compute,receive capital because they enable new revenue streams for model owners. This infrastructure buildout is essential for scaling always‑on agents and supporting the surge in daily active users reported by Meta. Consequently, investors are betting that the underlying platform layer will become as defensible and profitable as the AI models themselves.
Meta's Q3 results signal AI-driven growth amid rising costs
Meta reported $64.6 billion in revenue for Q3 2026, a 26.1 % year‑on‑year increase that beat analyst expectations by about 2 % Yahoo Finance. The beat was powered by higher user engagement, with daily active people rising 100 million to 3.64 billion, and by the early adoption of the Muse AI agent. Operating margin fell to 20.1 % from 40.1 % a year earlier, reflecting legal expenses, infrastructure build‑out, and the cost of launching new AI products. Management guidance of $74.5 billion for Q4 exceeds consensus, indicating confidence that AI investments will continue to translate into revenue.
The launch of Muse as a free, always‑on agent contrasts with OpenAI's Dots, which is limited to paying ChatGPT Pro subscribers, shaping early adoption curves. Competitors are also moving fast: Anthropic introduced Claude Duet for continuous background task execution, while OpenAI plans a $500 per month Pro tier and dedicated hardware. Meta's strategy of bundling AI with its social platforms gives it a unique distribution advantage, but it also raises questions about long‑term monetization beyond advertising. Regulatory scrutiny of high‑risk AI and the need for robust payment infrastructure for agent transactions add layers of uncertainty to the sector's outlook.
Meta's third quarter results showed that AI investments can drive both revenue and engagement, with sales reaching $64.6 billion and daily active people climbing to 3.64 billion. The launch of Muse and related AI features helped lift user activity, though the company still faced legal, infrastructure, and product costs. Meta's profit and revenue forecasts were stronger than many analysts expected, but operating margin fell sharply from the prior year. The quarter pointed to clear momentum, yet it also highlighted how expensive the AI era remains.
The next phase will likely test whether AI agents can scale without breaking cost controls or regulatory limits. Meta, OpenAI, and Anthropic are all pushing toward always-on agents, from Muse to Dots to Claude Duet, while enterprises rethink workflows. OpenAI's DevDay brought a paid rival to Meta's free agent, and Anthropic's research preview widened the debate about background execution. As these systems move deeper into commerce, productivity, and public policy, the hardest questions may be about oversight, spending, and accountability, or what happens when AI agents stop waiting for instructions and start deciding for themselves?
Frequently Asked Questions
How much revenue did Meta report in Q3 2026?
Meta reported $64.6 billion in revenue, up 26.1 percent from the same period a year earlier, and adjusted profit of $7.40 per share.
Why did Meta's operating margin decline?
The company cited legal expenses, infrastructure spending, and higher costs for emerging AI products, even as engagement grew.
What is Meta's Muse AI agent?
Muse is Meta's personal AI agent focused on shopping and productivity, and about 25 percent of people who set it up use it daily.
What did OpenAI announce at DevDay 2026?
OpenAI introduced Dots, a paid AI agent, announced the GPT-6.1 Sol model, and said ChatGPT reached 1.2 billion weekly users.
Is Anthropic's Claude Duet an always-on AI agent?
Anthropic announced Claude Duet as a research preview for Pro and Max users in the United States, designed to monitor apps and run tasks in the background.
