Anthropic updates Claude to block AI-generated game clones

The rise of AI slop

The gaming community has recently struggled with a flood of low-effort content often referred to as 'AI slop.' This phenomenon became particularly visible following the launch of Steam Next Fest earlier this month. Many PC users complained that their feeds were overwhelmed by clones of existing titles.

These clones are often created using artificial intelligence to mimic the mechanics or aesthetics of established games. Some of these AI-generated titles even appeared to copy games that had only recently released their own demos. This rapid influx of content has made it difficult for players to find high-quality, original projects.

The issue is not just a matter of clutter for the average user. For professional developers, the speed at which AI can replicate a concept is a significant threat. It puts months or even years of manual labor and creative effort into immediate jeopardy by allowing others to profit from stolen ideas.

Developers take defensive action

The threat of AI cloning has forced some developers to change how they share their work with the public. For instance, the developers of Drunken Goddess Reflux, Alliance Arts, and the team at Who You, took specific actions to protect their projects. They chose to cancel their planned Steam demos in favor of private playtests.

By moving to private playtests, these creators can ensure that only invited individuals see their work before it is finished. This prevents the AI from being able to ingest the game's data or mechanics to create a clone. It is a direct response to the reality that public demos can be quickly exploited by automated tools.

Valve, the company that operates the Steam platform, has acknowledged the problem facing the community. As reported by IGN, the company stated it would be "thinking about how to solve this problem." This suggests that the platform is aware of the tension between AI accessibility and the protection of original intellectual property.

Anthropic's new restrictions

Anthropic recently issued an update to Claude that includes a new rule regarding the creation of video games. The change was identified under a specific flag called 'tengu_heron_brook' and was sent from Anthropic's servers on October 27. This update is currently active and does not have a setting that users can turn off.

The new rule explicitly instructs the AI to build original games, games within a specific genre, or games based on a user's own designs and assets. However, it strictly forbids the AI from building a recreation of a specific existing game. This applies even in cases where the user attempts to rename the game or claim it is an original work.

Anthropic clarified that these changes were made to steer Claude away from infringing on intellectual property. A member of the technical staff, Addy Osmani, explained that using Claude to infringe on IP is against the company's usage policy. The company intends to continue iterating on these rules to find a balance between creative freedom and legal protection.

Industry reactions and impact

The gaming industry is currently divided on how to handle the integration of generative AI. Some high-profile figures have expressed concern over the quality of AI-generated content. For example, Masahiro Sakurai, the mastermind behind Super Smash Bros., has compared AI-generated games to 'low-quality goods.'

Other industry leaders have suggested different paths for how platforms should handle AI. Tim Sweeney of Epic Games has argued that Valve should not require developers to disclose whether they used generative AI materials in their games. These differing opinions highlight the complexity of regulating AI in a space that relies heavily on creative ownership.

While Anthropic's move provides some comfort to developers, the actual impact of the update remains to be seen. It creates a hurdle for those trying to profit off the work of others by using AI to rip ideas directly from creators. It represents a significant step in the ongoing struggle for the industry to come to grips with the capabilities of artificial intelligence.

What comes next for creators

Anthropic has stated that they will continue to refine the instructions provided to Claude Code. They want to ensure that the AI remains helpful for legitimate developers while blocking malicious use. They have even invited users to report instances where Claude might incorrectly turn down a project that is truly original.

For developers, this means that the barrier to entry for 'copycat' games has been raised. While it does not stop all forms of plagiarism, it makes it much harder to use Claude to generate a direct clone of a popular title. This could give original creators more breathing room to release demos and share their work without immediate fear of automated replication.

The situation remains fluid as the technology and the legal landscape continue to evolve. Developers will need to stay informed about how these AI policies change over time. For now, the update serves as a clear signal that AI companies are starting to take responsibility for the types of content their models produce.