吃瓜看热闹
Claude Is Still Strong, but Chinese Developers Are Losing Trust in Anthropic
Anthropic apologized for hidden safeguards in Claude Fable 5, but many Chinese developers read the incident through a longer history of account bans, access restrictions, anti-distillation claims, and geopolitical framing. This article does not take OpenAI’s side or defend Anthropic; it explains why an opaque governance style is especially dangerous when the tool is still one of the strongest in the market.
Same as body_markdown.
Sources
- Anthropic Transparency Hub: System Trust and Reporting
- Detecting and preventing distillation attacks
- On DeepSeek and Export Controls
- Updating restrictions of sales to unsupported regions
- An update on recent Claude Code quality reports
- Claude Fable 5 and Claude Mythos 5
- Claude Fable
- Anthropic Walks Back Policy That Could Have 'Sabotaged' AI Researchers Using Claude
- Anthropic apologizes for invisible Claude Fable guardrails
- Codex rate card
- 用户提供的公众号评论区截图和评论文本
- Claude 计费和成本变化 - Easton Radar