Anthropic Accidentally Publishes 512,000 Lines of Claude Code Without Security

Anthropic accidentally exposed 512,000 lines of Claude source code without security measures.

Published 12sem1 sourceImportant

≈ 18sSpeed

The fact

The incident reveals unclear protections on intellectual property of AI-generated code.

This flaw poses risks of intellectual property and non-consensual algorithm work appropriation.

Click the link to read an article on the topic:

Explore this topic

512 000 lignes de code Claude exposées publiquement sans contrôle

Auto-synthesis from 1 media source · identified on April 29, 2026