Google’s AI Rely on Facebook for Answers, Research Shows
New research reveals that Google’s AI Overviews frequently lean on social media platforms, particularly Facebook, to provide answers to user queries. Facebook was cited 19.5 million times, indicating a significant shift in how Google’s AI sources its information.
The Findings
- Social Media as a Primary Source: BrightEdge, an enterprise SEO firm, found that social platforms are now a core source for Google’s AI answers, with one in 15 searches being answered using social media content.
- Platform-Specific Expertise: Google’s AI treats different social media platforms as experts for various tasks. Facebook provides local and community signals, Instagram handles lifestyle and shopping queries, TikTok offers viral trends, Reddit provides firsthand experiences, and YouTube offers how-to context.
- Specific Examples: The study cites examples like a Cincinnati Reds Facebook post influencing a live sports query with an estimated 11.6 million monthly searches and an Instagram post leading to a mobile payment apps question.
Implications for Brands
The study suggests that for marketers, visibility on social media is no longer enough to ensure AI answers in favor of a brand’s website. Instead, brands need to focus on generating high-quality content that can influence AI responses, including posts from accounts they may not control. This phenomenon is part of a broader debate about generative engine optimization, citations in AI answers, and the economic implications of open web harvesting.
The Growing Scrutiny of AI
These findings come at a time when Google’s AI Overviews are under increasing scrutiny for accuracy and sourcing transparency. As AI continues to evolve, ensuring reliable and trustworthy information from these systems becomes increasingly important.