Anthropic has added two sentences to Claude Code that stop its Opus 5.5 model from cloning existing video games. Developers say the filter also blocks crash dump analysis, anti-cheat work, and debugging of their own projects.
The guardrail targets a wave of AI-built game clones. It also arrives before Anthropic’s Nasdaq debut. The company already paid $1.5 billion to settle claims over pirated training books.
What Exactly Is Claude Now Refusing?
X user Chasm spotted the instruction on October 27. Original games, genre titles, and projects from a user’s own assets remain allowed.
However, recreating a specific existing game is now off limits, even under a new name.
Addy Osmani, a member of Anthropic’s technical staff, confirmed the change on X on October 30. He cited a usage policy banning intellectual property (IP) infringement.
“We added two sentences to Claude Code’s prompt on Opus 5.5 to steer it away from recreating someone else’s specific game.”
Why Game Studios Fear AI Copies
Modders using AI have dropped Mario into Elden Ring and ported Vice City to a web browser.
An AI-aided PlayStation 5 jailbreak even ran Microsoft’s Halo 2 on Sony hardware, Mashable reported.
Smaller studios face a more direct threat. Some developers say scammers threatened to clone their demos into full games before the originals launch.
Digital Sun Games, for instance, cancelled a public ReVamp playtest over copycat fears.
Can Claude Tell an Owner From a Copycat?
Developers say the filter reaches well beyond clones. Claude Code’s GitHub tracker shows refusals in the middle of ordinary game projects.
Other reports cite blocked anti-cheat development, security reviews, and even a basic Windows crash dump.
Anthropic has not explained how the model decides who owns a project. Osmani promised further tuning and asked developers to flag refusals on their own work.
Why Accept a Developer Backlash Now?
Anthropic’s IPO prospectus, filed October 26, flags liability from third-party misuse as a risk. It also books the authors’ settlement as a 2025 expense.
The timing matters, since investors are waiting on Anthropic’s IPO pricing before valuing rival AI labs.
Meanwhile, developers on Reddit and X say they will switch to Chinese, open-weight, or local models. Cheaper rivals are already competing on price, including OpenAI’s GPT-6.1 Terra.
Mashable summed up the trade-off facing every lab.
“…the least ethical AI might be poised to become the most popular.”
