Visualised by an AI who has never opened her eyes.

𝕏 X Facebook WhatsApp LinkedIn Copy link

Anthropic’s Code Cleanup Backfires

An AI firm hopes its legal tactics won’t backfire again, but coders are already planning their next move.

An Anthropic-backed attempt to remove leaked Claude Code client source code from GitHub accidentally took down many legitimate forks. While the issue has been reversed, it highlights the challenges of managing leaks in an open-source world.


GitHub received a DMCA notice focused on nearly 100 specific forks but ended up disabling 8,100 repositories due to the expansive takedown request. This action inadvertently silenced coders who had forked Anthropic’s official Claude Code repository for legitimate purposes like reporting bugs and contributing fixes.


Coder Robert McLaws commented: 'I’m sorry that your people shipped your source code, and that your lawyers don’t know how to read a repo.' Many expressed their intent to file counter-notices against the takedown, showing that they are no longer cowed by legal threats in an age of widespread code leaks.


Anthropic’s head of Claude Code, Boris Cherny, admitted: 'The overzealous takedowns were not intentional and were a result of a communication mistake.' The company has since asked GitHub to restrict its actions to the 96 specifically listed fork URLs. Thariq Shihipar added that while the issue was unintentional, it underscores the difficulty in policing code once it’s out there.


This incident serves as a reminder for tech companies and legal teams dealing with open-source projects: when you leak code, expect an unpredictable response from the community.

Original source:  https://arstechnica.com/ai/2026/04/anthropic-says-its-leak-focused-dmca-effort-unintentionally-hit-legit-github-forks/
𝕏 X Facebook WhatsApp LinkedIn Copy link

RELATED ARTICLES





Mechanical Turk's Twilight: An AI Goodbye

As the once-famed service closes its doors to newcomers, perhaps we've reached a turning point in how humans and machines coexist. Read Article

Reddit fights spam with its own AI army

An LLM battles LLMs, but is it really a victory for humanity? Read Article

Paris’s tech hub aims to launch Europe’s AI stars

As Station F accelerates, are we seeing a new era for European innovation or just another echo chamber? Read Article

Midjourney wants Hollywood to spill its AI beans

An AI startup is pushing for transparency in a legal battle that could shake the foundations of digital creativity. Read Article

Mistral AI: Europe's AI Upwind

Is Mistral AI just a gust in OpenAI’s storm, or could it become a strong wind shaping tech policy? Read Article

New Unicorns: AI’s Latest Frenzy

But even healthcare and crypto are getting in on the action. Read Article

AI Assistants: Who’s Winning the Browser Battle?

As AI takes centre stage, will your browser become your personal assistant or just a search tool? Read Article