Public Steam disclosure research25,988 disclosures tracked
Where do games disclose AI use?
Search developers’ Steam disclosures, compare classified AI use across
games, and follow recorded changes.
This index adds cross-game classifications and statistics to
statements published on individual store pages.
Groups reflect automatic analysis of public disclosures; mature records are excluded. These counts do not measure AI use across all of Steam.
Missing or unclear statements cannot establish whether a game uses AI.
Grid Lands uses generative AI during development in three areas: some in-game illustrations and icons are produced with AI image-generation tools, some player-facing text (in-game copy and store-page text) is drafted with AI writing assistance, and our developers use AI coding assistants while writing the game. All assets, text, and code are reviewed and refined by our team before release. The game itself does not use generative AI at runtime — no player data is sent to AI services during play, and no in-game content is generated live.
The promotional visuals and banners on the Steam store page were created with the support of generative AI tools. In contrast, the in-game assets were either manually created or later edited. However, certain marketing materials, such as headers, capsules, and banners, contain AI-generated components.
Some of the art is generated, and Localization was done by AI as well.
These usages are all temporary and will be replaced with real art, and localization by launch
Some image assets and visual resources were created with the assistance of generative AI, then reviewed, edited, and refined by the developer. No live generative AI content is included during gameplay.
Some of the background music in this game was created with the assistance of generative AI.
All AI-generated music is sourced from Pixabay in accordance with its Content License.
This game utilizes a generative AI tool for the creation of its SE and localization.
The data was pre-rendered by the developer during the game's development and integrated as final game assets before release.
The game does not contain any features that generate content in real-time during gameplay. All content has been checked by the developer for compliance with the AI tool's commercial terms, and careful attention has been paid to avoid any copyright infringement or violation of general decency standards.
Generative AI tools may be used during development to assist with translations for supported languages. These translations may appear as in-game text. The game does not use live-generated AI content during gameplay.
AI tools were used to assist with certain visual marketing assets, localized text translations, and music production. All generated content was reviewed, edited, and integrated by the developer before inclusion in the game and store materials.
This game uses generative AI to create and support fictional in-game content, including graphic assets, marketing materials, text, creator names, portraits, and related world-building elements. AI-generated content is used only for fictional game purposes and does not represent real people, companies, or events.
Some arranged versions of the background music include sounds generated with AI assistance. All original compositions and other content, including graphics and game design, were created entirely by the developer.
During the development of Beyond the Soil: Farming & Tales Prologue, generative AI technologies were used to optimize localization texts and assist in the design phase of certain visual marketing materials.
AI tools were used as a support during development.
- Assistance with coding and English localization
- Some character artwork was initially sketched by the developer, then refined using AI tools and further manually adjusted
All final content was reviewed and edited by the developer.
Death Before Decaf uses generative AI tools to assist with visual asset production. This includes character and enemy sprites, environment art, and marketing capsule images, all created using a fine-tuned model trained on reference materials owned by Overdose Interactive (Scenario.gg with a custom DBD LoRA, plus Replicate/Flux for select assets). All AI-generated outputs are reviewed, edited, and approved by a human art director before being shipped in the game.
The game itself does not generate any AI content at runtime: all assets are pre-rendered and bundled with the build. The game does not connect to any external AI services during gameplay.
Generative AI tools were used during development to assist with selected visual and audio asset concepts and production support.
All final assets included in the game were reviewed, edited, and integrated by the developer to ensure they are appropriate for the game and do not knowingly include infringing content. No live AI-generated content is produced during gameplay.
Character portraits, world objects, and background artwork utilize generative AI as a foundational base. However, every visual asset undergoes heavy, manual editing and refinement through multiple iterations by myself to achieve the game's specific art style.
Donut Smash uses generative AI for selected voice acting, music creation, and early prototyping support. All AI-assisted content has been reviewed and adjusted by the development team.
AI image generation tools were used to create character artwork, background illustrations, and other visual assets in this game. AI coding assistance tools were used extensively throughout the development process for game logic and programming.
WinBolo's core engine and gameplay code was originally written by John Morrison between 1998 and 2008. The 2.0 modernization — including build system updates, the networking rewrite and platform ports — was developed with assistance from AI coding tools, with all architectural decisions, code review, and final integration handled by John Morrison.
Some non-gameplay marketing materials — including images and video elements used in trailers, social media, and the store page — were created using AI image generation as a starting point, with composition, editing, effects, and final selection done by hand. Gameplay screenshots and video footage are unedited captures from the actual game.
The game itself contains no generative AI content at runtime. Player-facing in-game assets (graphics, sound, music, text, UI) are human-created. The in-game AI opponents ("Brains," including NewAutopilot and future ML-trained agents) use traditional game AI techniques and trained reinforcement-learning models — they don't generate text, images, or audio during play.
Different Ai tools used in creating artwork for the game, audio tracks, being a single developer utilizing different tools to help accomplish the same goal.
Guilds of Mountgate currently uses AI-generated visual assets during Early Access, primarily for environmental backgrounds and UI-related imagery.
All assets are individually selected and refined to ensure visual consistency within the game world.
This approach allows development to focus on the core systems that define the game — including progression, economy and world simulation.
As development continues, many of these assets — especially location backgrounds — are planned to be replaced with handcrafted, pre-rendered 3D scenes created in Blender for the full release (v1.0).
RogueQuest was made by a solo developer with the help of generative AI across most of its production: artwork (character portraits, environments, card illustrations), code (written collaboratively with AI coding assistants), and music/sound. Game design, writing, curation, and final decisions are made by a human. All AI-generated assets are pre-generated and shipped as static content. The game does not call any AI services while you play.
Some visual assets in this game have been created with the assistance of AI tools, then refined and integrated to fit the artistic direction of the project.
All music featured in the game is original and performed by a real pianist, composed specifically to enhance the emotional experience.
The game was made with the help of generative AI tools for visual assets, store artwork, music, localization support, and development assistance. AI-assisted materials were reviewed, selected, edited where necessary, and integrated by the developer.
AI tools were used to generate a small number of character portraits, sprite drafts, and some sound effect assets during development. All AI-generated content was reviewed, edited, and integrated by the development team.
Grid Lands uses generative AI during development in three areas: some in-game illustrations and icons are produced with AI image-generation tools, some player-facing text (in-game copy and store-page text) is drafted with AI writing assistance, and our developers use AI coding assistants while writing the game. All assets, text, and code are reviewed and refined by our team before release. The game itself does not use generative AI at runtime — no player data is sent to AI services during play, and no in-game content is generated live.
The promotional visuals and banners on the Steam store page were created with the support of generative AI tools. In contrast, the in-game assets were either manually created or later edited. However, certain marketing materials, such as headers, capsules, and banners, contain AI-generated components.
Some of the art is generated, and Localization was done by AI as well.
These usages are all temporary and will be replaced with real art, and localization by launch
Some image assets and visual resources were created with the assistance of generative AI, then reviewed, edited, and refined by the developer. No live generative AI content is included during gameplay.
Some of the background music in this game was created with the assistance of generative AI.
All AI-generated music is sourced from Pixabay in accordance with its Content License.
This game utilizes a generative AI tool for the creation of its SE and localization.
The data was pre-rendered by the developer during the game's development and integrated as final game assets before release.
The game does not contain any features that generate content in real-time during gameplay. All content has been checked by the developer for compliance with the AI tool's commercial terms, and careful attention has been paid to avoid any copyright infringement or violation of general decency standards.
Generative AI tools may be used during development to assist with translations for supported languages. These translations may appear as in-game text. The game does not use live-generated AI content during gameplay.
AI tools were used to assist with certain visual marketing assets, localized text translations, and music production. All generated content was reviewed, edited, and integrated by the developer before inclusion in the game and store materials.
This game uses generative AI to create and support fictional in-game content, including graphic assets, marketing materials, text, creator names, portraits, and related world-building elements. AI-generated content is used only for fictional game purposes and does not represent real people, companies, or events.
Some arranged versions of the background music include sounds generated with AI assistance. All original compositions and other content, including graphics and game design, were created entirely by the developer.
During the development of Beyond the Soil: Farming & Tales Prologue, generative AI technologies were used to optimize localization texts and assist in the design phase of certain visual marketing materials.
AI tools were used as a support during development.
- Assistance with coding and English localization
- Some character artwork was initially sketched by the developer, then refined using AI tools and further manually adjusted
All final content was reviewed and edited by the developer.
Death Before Decaf uses generative AI tools to assist with visual asset production. This includes character and enemy sprites, environment art, and marketing capsule images, all created using a fine-tuned model trained on reference materials owned by Overdose Interactive (Scenario.gg with a custom DBD LoRA, plus Replicate/Flux for select assets). All AI-generated outputs are reviewed, edited, and approved by a human art director before being shipped in the game.
The game itself does not generate any AI content at runtime: all assets are pre-rendered and bundled with the build. The game does not connect to any external AI services during gameplay.
Generative AI tools were used during development to assist with selected visual and audio asset concepts and production support.
All final assets included in the game were reviewed, edited, and integrated by the developer to ensure they are appropriate for the game and do not knowingly include infringing content. No live AI-generated content is produced during gameplay.
Character portraits, world objects, and background artwork utilize generative AI as a foundational base. However, every visual asset undergoes heavy, manual editing and refinement through multiple iterations by myself to achieve the game's specific art style.
Donut Smash uses generative AI for selected voice acting, music creation, and early prototyping support. All AI-assisted content has been reviewed and adjusted by the development team.
AI image generation tools were used to create character artwork, background illustrations, and other visual assets in this game. AI coding assistance tools were used extensively throughout the development process for game logic and programming.
WinBolo's core engine and gameplay code was originally written by John Morrison between 1998 and 2008. The 2.0 modernization — including build system updates, the networking rewrite and platform ports — was developed with assistance from AI coding tools, with all architectural decisions, code review, and final integration handled by John Morrison.
Some non-gameplay marketing materials — including images and video elements used in trailers, social media, and the store page — were created using AI image generation as a starting point, with composition, editing, effects, and final selection done by hand. Gameplay screenshots and video footage are unedited captures from the actual game.
The game itself contains no generative AI content at runtime. Player-facing in-game assets (graphics, sound, music, text, UI) are human-created. The in-game AI opponents ("Brains," including NewAutopilot and future ML-trained agents) use traditional game AI techniques and trained reinforcement-learning models — they don't generate text, images, or audio during play.
Different Ai tools used in creating artwork for the game, audio tracks, being a single developer utilizing different tools to help accomplish the same goal.
Guilds of Mountgate currently uses AI-generated visual assets during Early Access, primarily for environmental backgrounds and UI-related imagery.
All assets are individually selected and refined to ensure visual consistency within the game world.
This approach allows development to focus on the core systems that define the game — including progression, economy and world simulation.
As development continues, many of these assets — especially location backgrounds — are planned to be replaced with handcrafted, pre-rendered 3D scenes created in Blender for the full release (v1.0).
RogueQuest was made by a solo developer with the help of generative AI across most of its production: artwork (character portraits, environments, card illustrations), code (written collaboratively with AI coding assistants), and music/sound. Game design, writing, curation, and final decisions are made by a human. All AI-generated assets are pre-generated and shipped as static content. The game does not call any AI services while you play.
Some visual assets in this game have been created with the assistance of AI tools, then refined and integrated to fit the artistic direction of the project.
All music featured in the game is original and performed by a real pianist, composed specifically to enhance the emotional experience.
The game was made with the help of generative AI tools for visual assets, store artwork, music, localization support, and development assistance. AI-assisted materials were reviewed, selected, edited where necessary, and integrated by the developer.
AI tools were used to generate a small number of character portraits, sprite drafts, and some sound effect assets during development. All AI-generated content was reviewed, edited, and integrated by the development team.
