Meta Business Terms Update Implies Consent for AI Data Scraping

Recent changes to Meta's Business Terms of Service have significant implications for how user data can be used. According to legal analysis, users with business accounts are now effectively consenting to their data being scraped for AI training purposes. This change coincides with the removal of previous clauses that addressed scraping and AI training, creating a new default for data usage.

What Changed in the Terms

The core issue stems from updates in the legal framework governing Meta's business accounts. As reported by sources, the current terms are interpreted as granting consent for AI scraping. This interpretation is reinforced by a separate finding that Meta has removed language that previously caused concern regarding the use of user data for scraping and AI training. For businesses operating on Meta's platforms, this means that their content and data may be used to train Meta's AI models.

Broader Context: New AI Model Availability

These policy changes occur as Meta expands its AI offerings. The company recently announced that its Muse Spark Open model is now available through Amazon SageMaker JumpStart. Muse Spark Open is described as an open-weight, frontier multimodal mixture-of-experts model, indicating Meta's continued development and deployment of advanced AI technologies that require large datasets for training. The availability on a major cloud platform like AWS makes this technology more accessible to developers.

Additional sources: - techcrunch.com - huggingface.co