Anthropic发布Claude 4，专注于企业安全与合规需求。
Anthropic launched Claude 4 in June, positioning it as an enterprise-grade AI assistant. The model supports an ultra-long context window (100K tokens) and features built-in data encryption and audit logs, complying with GDPR and CCPA. Claude 4 excels at enterprise tasks like contract analysis and compliance review, with 30% lower error rates than GPT-5. Companies like JPMorgan and Siemens are currently trialing it.