Anthropic admits partial leak of Claude Code source, says no customer data exposed

Anthropic confirms Claude Code source code leak, says no customer data exposed, calls it human error and outlines new safeguards after recent data lapses

Anthropic has confirmed that part of the internal source code for its artificial intelligence coding assistant Claude Code was leaked, with the disclosure coming on Tuesday, according to a report by CNBC.

The company stated that no sensitive customer data or credentials were involved or exposed, adding that the incident resulted from a release packaging issue caused by human error rather than a security breach, and that measures are being rolled out to prevent a recurrence, as per the statement cited by CNBC.

The Claude source code leak is seen as a setback for the startup, as it could provide software developers and competitors with insights into how the company built its coding tool, with a post on X containing a link to the code drawing more than 21 million views after being shared at 4:23 a.m. ET on Tuesday, according to reports.

The incident marks Anthropic's second major data lapse within a week, following a report by Fortune which stated that descriptions of the company's upcoming AI model and other documents were discovered in a publicly accessible data cache.

Anthropic, founded in 2021 by a group of former OpenAI executives and researchers, is best known for developing the Claude family of AI models. The company released Claude Code to the general public in May, with the tool designed to help software developers build features, fix bugs and automate tasks.

Claude Code has seen significant adoption over the past year, with its run-rate revenue exceeding $2.5 billion as of February, according to reports. Its growth has prompted companies including OpenAI, Google and xAI to invest in developing competing offerings in the AI coding tools segment.