Anthropic has clarified its data usage policies, particularly concerning user feedback provided through its chat interface. For commercial products like Claude for Work and the Anthropic API, user inputs and outputs are not used for model training by default. However, if users explicitly provide feedback via the thumbs up/down button or opt-in to data sharing, their conversations may be used for training, analysis, and research. This feedback data is de-linked from user IDs and stored for up to five years, with options for administrators to disable feedback submission for their organizations. AI
IMPACT Clarifies data privacy for users of commercial AI chat products, impacting user trust and adoption.
RANK_REASON Article explains data usage policy for a commercial product, not a new release or research.
Read on HN — anthropic stories →
- Anthropic
- Anthropic API
- Claude Code
- Claude for Work
- Claude Free
- Claude Gov
- Claude-Max
- Claude Pro
- Google Drive
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
