option
Home
Flash News
Content
KennethRoberts
KennethRoberts
April 1, 2026

AI firm Anthropic accidentally exposed the source code of its new command-line tool due to a publishing error, revealing nearly 2000 files and over 500000 lines of TypeScript. The leaked code includes unreleased features like cyber pets and an always-on AI with a nightly training mechanism. The incident highlights significant security risks in AI development.

AI firm Anthropic accidentally exposed the source code of its new command-line tool due to a publishing error, revealing nearly 2000 files and over 500000 lines of TypeScript. The leaked code includes unreleased features like cyber pets and an always-on AI with a nightly training mechanism. The incident highlights significant security risks in AI development.
Comments (0)
0/300
OR