Archive — US — techcrunch
Bluesky users debate plans around user data and AI training
Bluesky, a social network, proposed user data options for AI training and archiving, causing user backlash over perceived shifts in data privacy. CEO Jay Graber defended the plan, stating it aims to set a standard for AI scraping. Users can opt-out of data usage in four categories. Critics argue that these consent signals depend on ethical compliance from scrapers, which isn't legally enforceable.
Key points
- Bluesky issued a proposal about user data permissions on GitHub.
- Users can indicate their preferences regarding data scraping for AI and archiving.
- CEO Jay Graber emphasized the need for a new scraping standard due to existing public data scraping.
- The proposed system is likened to robots.txt, offering a non-legally enforceable ethical guideline.
- User reactions were mixed, with some feeling betrayed over potential data usage for AI despite previous reassurances.
Why it matters
- This reflects a growing trend in tech where platforms are pressured to clarify data usage in AI contexts, alongside privacy concerns.
- Historically, similar debates occurred around data scraping and usage, especially after GDPR implementation, highlighting user consent.
- By 2025, IT trends may increasingly prioritize transparent data usage policies amidst AI's expansion.
Bluesky's proposal raises significant questions about user consent and data privacy in the age of AI, demonstrating the need for clear standards and robust ethical practices that align with user expectations.