Meta is preparing to turn its Muse AI agent into a new revenue stream through commerce transaction fees, paid subscriptions and business services. The strategy was outlined by CEO Mark Zuckerberg and CFO Susan Li during Meta’s third-quarter 2026 earnings call, as the company discussed how it intends to monetise its growing AI business. Instead of relying exclusively on advertising, Meta wants Muse to help users discover products, complete transactions and manage work while generating revenue from the services surrounding those activities.
The plan builds on Muse’s early adoption. More than 10 million people in the United States and Canada had set up the AI agent within approximately six weeks of its launch, according to Zuckerberg. Shopping has emerged as a major use case, alongside personal productivity and creative work. Meta expects most users to access core features for free, while businesses, advanced users and companies seeking specialised AI capabilities could become paying customers.
Key Takeaways
- Commerce fees: Meta plans to charge businesses fees for transactions that Muse helps facilitate.
- Paid subscriptions: Users will be able to pay for advanced AI models and additional usage.
- Business services: Small businesses and enterprises could subscribe for specialised features and workplace applications.
- Shopping opportunity: More than a quarter of Muse users engage with shopping, according to Meta.
- Free core access: Meta expects the main experience to remain free for most consumers.
- AI infrastructure costs: The company aims to improve computing cost efficiency by more than tenfold over the next year.
- Long-term goal: Meta wants to connect consumer and business AI agents, creating new ways to generate revenue from digital transactions.
Meta’s Three-Part Muse Monetisation Strategy
Meta’s proposed business model has three primary components: transaction fees, subscriptions and services for businesses. Each targets a different source of revenue and could allow the company to earn money from activities that extend beyond conventional advertising.
1. Commerce fees from AI-assisted purchases
The first opportunity is earning fees from transactions that Muse helps facilitate. Rather than charging every consumer to use the assistant, Meta expects businesses to pay fees for commerce tools that help users complete purchases.
For example, a consumer could ask Muse to research products, compare options and help complete a supported transaction. If the agent facilitates a purchase, Meta could earn a fee under the relevant commercial arrangement.
The model would give Meta a way to participate more directly in online commerce. Its existing social platforms are already important channels for product discovery, advertising and brand engagement. Muse could extend that relationship into the purchasing process itself.
However, the company has not disclosed a universal transaction-fee rate or a detailed schedule of merchant charges. The actual revenue opportunity will depend on how many purchases flow through the agent, the merchants participating and the terms of its commercial arrangements.
2. Subscriptions for advanced AI features
The second revenue stream involves subscriptions for users who want more than the standard free experience.
Meta expects users seeking its most advanced AI models or additional usage to be able to pay directly. This could include people who use AI frequently for research, planning, content creation or other demanding tasks.
The approach resembles a broader model used across the AI industry: offer a free entry point to attract users, then charge for higher usage limits or specialised capabilities.
For Meta, the challenge will be determining which premium features users consider valuable enough to pay for. Subscription revenue will depend on repeat engagement, perceived improvements over the free version and the cost of providing more intensive AI services.
3. Business and enterprise services
The third component targets small businesses and larger organisations. Meta wants Muse to help businesses acquire customers, manage operations and improve productivity, while offering additional features through paid subscriptions.
The company has already introduced Muse for Small Business, which connects with tools such as Shopify, QuickBooks, Canva, Slack, Stripe and other business applications. These integrations can help the agent work with information about a company’s products, customers, marketing and operations.
Potential uses include analysing advertising performance, drafting customer responses, organising direct messages, preparing marketing campaigns and identifying operational issues.
Meta has estimated that at least 500 million self-identified small businesses use its family of apps to reach customers and generate revenue. That represents a large potential audience, although it does not mean all those businesses will adopt Muse or pay for its services.
How Muse’s Early Shopping Activity Supports the Plan
Shopping has become one of Muse’s most important early use cases. During Meta’s third-quarter earnings call, Zuckerberg said more than a quarter of Muse users were engaging with shopping and that the average purchase value was above $200.
These figures give Meta an early indication that people are willing to use the agent for commercial tasks, not just general questions. They also help explain why transaction-based monetisation is central to the company’s strategy.
| Metric | What Meta reported |
|---|---|
| Muse adoption | More than 10 million setups in the US and Canada in roughly six weeks |
| Shopping engagement | More than a quarter of users engage with shopping |
| Average purchase value | Above $200 |
| Main monetisation routes | Commerce fees, subscriptions and business services |
| Core consumer experience | Expected to remain free for most users |
| Computing efficiency target | More than tenfold improvement over the following year |
Note: These are company-reported figures and targets. They do not establish Muse’s current profitability or the amount of revenue generated directly by the agent.
The distinction between purchase value and Meta revenue is important. An average purchase above $200 does not mean Meta earns $200 per transaction. Its potential income would depend on the fees charged, the volume of transactions and the commercial terms agreed with businesses.
Why Meta Wants to Move Beyond Advertising
Advertising remains Meta’s core business, but Muse gives the company a potential way to participate in more parts of the digital customer journey.
Facebook and Instagram help users discover products through posts, creators and advertisements. An AI agent could extend that journey by helping people identify what they want, compare alternatives and act on a purchasing decision.
That shift could create new commercial opportunities. Businesses might pay Meta not only to reach potential customers through advertising but also to access tools that support sales, customer service and transaction completion.
AI agents could also become useful workplace tools. If Muse helps businesses manage routine tasks, prepare campaigns or coordinate customer communications, Meta could generate subscription revenue from the productivity benefits it provides.
The strategy would diversify Meta’s monetisation options, but it does not mean advertising will become less important. Instead, AI could strengthen the existing advertising business while opening additional revenue streams.
Computing Costs Are a Major Challenge
Building a popular AI agent is only part of the challenge. Running it at scale can be expensive, particularly when users ask it to perform complex tasks across multiple applications.
Meta says Muse is computationally intensive and that serving costs will rise as the product expands. Zuckerberg said the company aims to improve computing cost efficiency by more than tenfold over the next year through more efficient models, better use of available computing capacity and inference optimisation.
Such improvements would be important to the business model. If each interaction becomes cheaper to serve while subscriptions and commerce revenue grow, Muse could develop more sustainable economics.
However, the tenfold improvement is a company target, not an already demonstrated outcome for the entire business. Meta has not provided a complete breakdown of Muse’s standalone revenue, operating costs or profitability.
The company is therefore attempting to solve two problems simultaneously: expanding adoption and improving the economics of serving a large number of users.
Competition and Consumer Trust Could Shape Growth
Meta is not the only company developing AI assistants that can act on users’ behalf. OpenAI, Anthropic and other technology firms are also working on systems that connect conversational AI with software, shopping and business workflows.
The competition will depend on more than the underlying AI model. Integration quality, reliability, transaction security, pricing and ease of use will all influence whether consumers and businesses adopt these services.
Retailer participation is another potential obstacle. Businesses may welcome additional sales opportunities but remain concerned about losing control over customer relationships, product discovery and purchase data. Some retailers may restrict access to their platforms or require specific commercial agreements.
Trust will also matter. Consumers need to understand what an AI agent can access, what it can purchase and when it requires approval. Businesses need confidence that the system will handle customer information appropriately and avoid unauthorised actions.
Meta’s ability to establish clear safeguards and commercially attractive arrangements could influence how quickly Muse’s transaction-based business develops.
The Bigger Picture
Meta’s Muse strategy represents an attempt to extend its business model from advertising and social engagement into AI-assisted commerce and productivity. Transaction fees could allow the company to earn revenue when its agent helps complete purchases, while subscriptions and business services could monetise advanced capabilities. The approach also gives Meta an opportunity to build on its existing relationships with consumers, advertisers and small businesses.
The commercial outcome remains uncertain. Muse must retain users, generate repeat transactions, attract paying customers and operate at a sustainable cost. Meta’s planned improvements in computing efficiency will be particularly important because growing usage without controlling serving costs could put pressure on profitability. The company has outlined several possible revenue streams, but their contribution will become clearer only as pricing, transaction volumes and business adoption develop.
FAQs
1. How does Meta plan to make money from Muse?
Meta plans to generate revenue through fees on transactions the agent helps facilitate, subscriptions for advanced AI features and additional services for businesses and enterprises.
2. Will Meta Muse remain free?
Meta expects the core experience to remain free for most users. People who want advanced models or additional usage, as well as businesses seeking specialised features, may pay for subscriptions.
3. How will commerce fees work?
Meta expects businesses to pay fees for commerce tools that help Muse facilitate transactions. The company has not announced a universal fee rate or a complete pricing structure.
4. How can businesses use Muse?
Businesses can connect Muse with supported applications to analyse performance, prepare marketing campaigns, organise customer messages and manage aspects of their operations. More advanced features may be offered through paid services.
5. Is Meta Muse profitable yet?
Meta has not disclosed enough standalone financial information to establish Muse’s profitability. The company has said computing costs are substantial and that it is targeting major efficiency improvements as the service scales.
Looking Ahead
Meta’s immediate challenge is to turn Muse’s early adoption into repeat usage and sustainable revenue. Commerce fees, subscriptions and business services provide several possible paths, but each requires users and businesses to see enough value to keep using the agent. The number of purchases facilitated, the rate of paid subscription adoption and the performance of business tools will be important indicators of progress.
Over the longer term, Meta wants consumer and business agents to connect and transact directly, potentially creating a new layer of AI-driven commerce. Achieving that vision will require reliable integrations, retailer cooperation, strong user protections and lower computing costs. Muse could become a meaningful extension of Meta’s existing business, but the scale of its contribution will depend on execution and the economics of AI services.
Get the day’s top stories in your inbox
One concise email. No spam, unsubscribe anytime.
