Anthropic accidentally leaked over 500,000 lines of Claude code
Anthropic, a U.S.-based AI company, accidentally leaked over 500,000 lines of code from its Claude Code tool, and Chinese developers wasted no time jumping on it.
Even though Anthropic blocks its services in China, the leaked code was reposted on GitHub, sparking major buzz in developer circles.
Chinese forums dissect Claude code
Chinese forums are now full of discussions about Claude Code's advanced features and design choices.
Many see the leak as a rare peek into Anthropic's engineering secrets: Beijing IT architect Zhang Ruiwang even called it
But the code batches are indeed a treasure for AI companies or developers, as they revealed all the key engineering decisions Anthropic made.
While some industry experts claimed the leak included Claude Code source code but not the model weights, this leak still raises big questions about tech security and ongoing U.S.-China competition in AI.