Meta (formerly Facebook)
The world's largest social network company (Facebook, Instagram, WhatsApp). Through its open-source LLaMA model series, it leads the open-source AI ecosystem, while exploring AI interaction entry points in the physical world via Ray-Ban smart glasses.
Its social network moat is unshakable, with an advertising business generating extremely strong cash flow from billions of users worldwide. The open-source LLaMA models have become the default choice for AI developers, using an open-source strategy to undercut the market positioning of closed-source model vendors.
Company Overview
Meta Platforms (formerly Facebook) was founded in 2004 by Mark Zuckerberg in his Harvard dorm room. Over the past 20 years, the company has undergone three major transformations: from a college social network to a global social empire, and now to an all-in bet on AI and the metaverse.
Since 2023, Meta has gone from an AI "follower" to a leader of the open-source AI ecosystem through its open-source LLaMA (Large Language Model Meta AI) series. Unlike OpenAI and Google, which have chosen closed-source routes, Meta's strategy is to "make LLaMA the Linux of the AI field through open source," thereby dismantling competitors' business models while feeding AI capabilities back into its own social products and advertising systems.
Core Business
Open-Source Large Models (LLaMA Series) | Strategic Core
- LLaMA 2: A free, commercially usable open-source large model released in 2023, becoming one of the most widely used foundation models in the AI field.
- LLaMA 3 / 3.1: The 8B, 70B, and 405B parameter versions released in 2024, reaching or even surpassing GPT-3.5-level performance across multiple benchmarks. The 405B is currently the largest open-source model in the world.
- Strategic Logic: Meta does not directly charge for its models. Instead, through the widespread adoption of LLaMA, it unifies the innovation of the open-source community and third-party developers onto Meta's technology stack, ensuring Meta is not excluded from the ecosystem in the AI era.
Social Network Advertising (Core Revenue Source, Over 98% of Revenue)
Facebook and Instagram have more than 3 billion daily active users. AI technology is deeply embedded in the advertising system:
- AI Recommendation Engine: Deep-learning-based content recommendation algorithms determine what users see in Feeds and Reels, directly impacting user retention and ad monetization efficiency.
- GenAI Advertising Tools: AI automatically generates ad creatives, images, and videos, significantly lowering advertisers' content production costs.
Ray-Ban Meta Smart Glasses
Meta's most important hardware-side AI exploration. Built in partnership with Ray-Ban, the smart glasses are equipped with a camera, speakers, and an AI assistant, supporting first-person capture, real-time object recognition, and voice interaction. Shipments exceeded one million pairs in 2024, making them one of the world's most successful AI wearable devices.
Technical Moat
Open-Source Ecosystem Leadership: Meta has gained broad support from the developer community and ecosystem building through open-sourcing LLaMA. This open-source strategy effectively weakens the closed-source barriers of OpenAI and Google, positioning Meta on the side of developers as a "good citizen."
Social Network Data Flywheel: The social graphs, content preferences, and interaction behaviors of billions of users form an unmatched AI training data pool. These data are not only used to optimize advertising algorithms, but also provide LLaMA models with a unique understanding of social context.
Market Landscape
| Dimension | Data |
|---|---|
| Open-Source Large Models | LLaMA series is the de facto standard |
| Social Media DAU | Over 3 billion |
| Smart Glasses Shipments | Ray-Ban Meta exceeds one million pairs |
| Key Competitors | ByteDance/TikTok, Google, OpenAI |
Financials and Growth
| Metric | Data |
|---|---|
| Revenue (2024) | Approximately $160 billion |
| Net Income | Approximately $47 billion |
| AI Capital Expenditure | Approximately $35 billion |
| Growth Drivers | Advertising business growth + AI recommendation efficiency gains |
Risks and Summary
Key Risks:
- Short-term pressure on profits from massive AI investment — Meta's capital expenditure has expanded significantly in AI infrastructure.
- Improper use of open-source models could lead to public relations and legal risks.
- TikTok's sustained competition poses a threat to Meta's social media core business.
Core Industry Value: Meta is the "guardian" of the open-source AI ecosystem. Its open-source strategy effectively prevents large models from being monopolized by a few closed-source companies and drives the democratization of the entire AI industry. On the hardware front, Ray-Ban Meta glasses demonstrate the enormous market potential of AI wearables.