House House has publicly described its use of language models during Big Walk's development and acknowledged that it had no internal AI-use policy. For players and other studios, that admission makes the future rules for development as relevant as the amount of code attributed to tools.
The current official FAQ sets out the account. Kotaku reported it on November 10, 2026; the FAQ section does not display a separate update date, so the report's date should not be treated as the precise moment the statement was added.
A code percentage answers a narrow question
The studio estimates that substantially less than 0.1% of the game's code was connected to this usage. It describes chatbots helping investigate technical problems, with some suggestions modified into code, and says generative AI was not used in other parts of making the game.
That is the developer's account, not a published independent audit. The percentage describes a code-volume estimate; it cannot measure the creative importance of a particular change, the provenance of every suggestion or how individual players weigh the practice.
Readers should therefore separate two judgments: what the studio says happened, and whether that process meets their own expectations. A small reported quantity does not settle a disagreement about whether a tool should have been used at all.
A store disclosure and a public explanation serve different purposes
House House says it first prepared a version of the disclosure for its IGF submission. It explains its lack of a Steam disclosure by saying the platform does not request disclosure of code or chatbot use.
Valve's current Content Survey documentation focuses its generative-AI questions on content delivered to and consumed by players, including artwork, audio, narrative and localization. It distinguishes development-tool efficiency from that focus.
This helps explain the scope of the survey. It does not make a missing storefront label proof that no AI-assisted work occurred, and it does not make platform compliance a complete statement of a studio's creative policy. Those are different questions for different audiences.
The promised policy needs concrete boundaries
The FAQ calls the absence of an internal policy a failure and says the studio will work toward a more comprehensive one. That is a commitment to further work; it does not yet specify the finished rules.
A useful policy would explain which activities are permitted, who reviews exceptions and how tool use is recorded and communicated. Those are editorial suggestions for clarity, not requirements announced by House House.
The next meaningful evidence will be the policy itself and how the studio describes applying it. The current statement gives players a more specific account to assess without resolving every ethical disagreement or establishing an awards decision.
Sources
Related Articles
Xbox PC app preview targets the wait between opening the launcher and playing
The November 5 Insider update reorganises the library and social tools. Microsoft's loading-time claims describe the app, with a separate rollout for cloud screens.
Macronix’s ROM phase-out: what the figures establish for Switch game cards
ROM fell to 5% of Macronix’s parent-company product sales. That shift and Nintendo’s existing game-key cards answer different questions about physical games.
Nexus Mods sets January 2 deadline for existing mods' new AI disclosures
New uploads must complete the updated fields immediately. A clarified transition period gives existing authors time for detailed AI-assisted statements and permission settings.
Dead of the Brain reaches Steam with English text and optional hints
FAKKU's rerelease adds official language access and usability options to the 1992 PC-98 adventure. Its older interface remains part of the experience.