TL;DR
- Meta CEO Mark Zuckerberg says AI is increasing user engagement across Facebook and Instagram.
- Time spent rose by 5% on Facebook and 6% on Instagram, due to better AI recommendations.
- 3.4 billion daily active users across Meta’s app ecosystem in June 2025.
- Video watch time rose 20% year-over-year, supported by ranking system improvements.
- Threads saw improved engagement driven by large language model (LLM) integration.
AI as a Growth Catalyst
During Meta’s Q2 2025 earnings call, CEO Mark Zuckerberg emphasized the role of artificial intelligence in driving stronger engagement across the company’s platforms.
“AI is significantly improving our ability to show people content that they’re going to find interesting and useful,” said Zuckerberg.
Thanks to advancements in AI-driven recommendation algorithms, time spent on Facebook increased by 5% and Instagram usage grew by 6% during the quarter. This rise comes despite growing public concern about the flood of “AI slop” — low-quality content generated by artificial intelligence — across social platforms.
Meta’s Expanding Ecosystem
Meta reported that its daily active user base across its family of apps — including Facebook, Instagram, Messenger, and WhatsApp — reached 3.4 billion in June 2025, a 6% increase from the previous year.
This increase helped push family of apps revenue to $47.1 billion, a 22% year-over-year gain. Much of this growth was attributed to better AI content delivery and user engagement across platforms.
Video and Threads Gain Momentum
The company also highlighted a 20% year-over-year increase in video watch time, citing improvements in content ranking systems and an emphasis on original video content on Instagram Reels. These optimizations are part of Meta’s broader strategy to compete with TikTok and retain younger audiences.
Meta’s Twitter alternative, Threads, is also gaining traction. The platform has seen time spent gains driven by the integration of large language models (LLMs), reflecting Meta’s broader investment in generative AI for content recommendations and creation.
Key Engagement Metrics (Q2 2025)
| Metric | Value | Source |
| Daily active users (Meta apps) | 3.4 billion | Meta |
| Revenue (Family of Apps) | $47.1 billion (↑22% YoY) | Meta |
| Time spent on Facebook | ↑ 5% | Meta Earnings Call |
| Time spent on Instagram | ↑ 6% | Meta Earnings Call |
| Video time spent (Instagram) | ↑ 20% YoY | TechCrunch |
| Threads engagement | Boost from LLM integration | Meta |
Meta’s Bet on AI Is Paying Off — For Now
While user sentiment around AI-generated content remains mixed, Meta’s Q2 results suggest its AI-centric strategy is effective — at least in terms of engagement and monetization. Zuckerberg has consistently framed AI as a central pillar of Meta’s future, reinforcing that the technology is improving content discovery, platform stickiness, and ultimately, ad revenue.
Meta’s continued push to integrate AI in products like Meta AI and smart glasses like the Ray-Ban Meta reflects this broader ambition. Meanwhile, the success of Threads, now fueled by language model-driven enhancements, positions Meta as a strong player in the AI-powered social media race.
Still, balancing quality content with AI-driven scalability will be key as Meta looks to sustain long-term growth while managing public and regulatory scrutiny over AI moderation and transparency.
