Public Steam disclosure research25,497 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.
Tumble's visual art (character and charm illustrations, backdrops, item sprites, and store capsule art) was generated with Google's Gemini image models and then selected, cropped, retouched, colour-matched and in many cases regenerated by hand against a fixed reference style until it fit the game. Music and sound effects were generated with Suno under a commercial licence, then cut, mixed and placed by hand.
The game's design, rules, balance, and code are not generated content. Every mechanic, number and line of player-facing text was written, tested and revised by the developer, including through automated balance simulation and four rounds of human playtesting.
No content is generated live during play. Nothing in Tumble is produced at runtime by an AI model; the game runs entirely offline with no model calls of any kind.
AI has been used solely as a coding assistant for assistance purposes.
As a solo indie developer with no software background, using a coding assistant is my only option.
Regards.
We used Artificial Intelligence tools as an aid during development and in the creation of some game content. All AI-assisted material was personally reviewed and adjusted by our team, who were involved throughout every stage of the creative process.
本作在开发阶段使用了生成式 AI 辅助制作以下内容：美术（场景与背景、卡面插画、角色头像、界面位图、物品图标等游戏与商店页中的图像素材）由 AI 生成母图、经开发者逐项筛选、修复、裁切与组装后定稿；游戏内文本（事件与叙事文案）由开发者设计框架、AI 辅助起草、逐条经人工审查修改后定稿。游戏运行过程中不会实时生成任何 AI 内容，也不会连接任何 AI 服务。核心玩法、世界设定、事件结构与全部最终内容取舍均由开发者完成。
Final Night Firefight uses generative AI tools during development to assist with the creation of some pre-generated game artwork, visual assets and music. I review, edit, and integrate all AI-assisted content as part of the game's development process. The game does not generate AI content live during gameplay.
Generative AI was used to create the game's English and Simplified Chinese localization. AI tools were also used to assist with code generation during development. No AI-generated content is created during gameplay.
Some visual and audio assets in this game were created with the assistance of generative AI tools, and subsequently reviewed, edited, and refined by the developer. The game does not feature live or runtime generative AI during gameplay and does not generate content based on player input.
AI tools were used during the creation of draft visual assets. All AI-generated material is reviewed, edited, and manually refined by professional artists before being included in the final product.
Generative AI tools were used in the creation of certain assets for Camedieo.
Each was reviewed, reworked and integrated by us so the finished product holds a single consistent direction.
Generative AI tools were used during development to assist with the creation and iteration of some some pre-generated visual, written and promotional assets. All AI-assisted content is reviewed, edited and curated by the developer. No generative AI is used during gameplay.
Generative AI was used during development to assist in creating some 2D art assets, including scene backgrounds, character and monster artwork, UI elements, and icons for items, equipment, skills, and status effects. Generative AI was also used to assist in drafting some in-game text. All AI-assisted content was selected, edited, post-processed, and integrated by the developer. The game does not use generative AI at runtime and does not generate content in real time in response to player input.
Pocket Pawns was made with the help of generative AI, and the game's own credits name every tool used.
Images and music: the chess piece art was generated with ChatGPT, the board art with Gemini, and the world and environment art from assets created with Nano Banana Pro alongside free-to-use CC0 packs found online. The soundtrack is from Suno.
Claude wrote the game's code, its writing, and its translations into ten languages — and made much of what you see and hear directly in code rather than from an image generator: the character pixel art, the in-world props and interface, the three-dimensional rooms, the animation, and the sound effects.
The chess itself is not generated. The puzzles come from the Lichess puzzle database, released under CC0, and your opponent is Stockfish, a classical open-source chess engine. Real positions, a real engine.
Nothing is generated while you play. Pocket Pawns creates no content at runtime and never connects to an AI service. It does not need an internet connection to play.
Crie au loup is made with AI, from the first prototype to the illustrations. That is not a manufacturing detail: it is as much the project as the game itself.
My role is not a craftsman's: I painted no card, and I did not write the code line by line. I designed the system, I direct it, I arbitrate and I verify. Deciding what makes it into the game, what gets redone, what gets cut: that is production work, and it is mine.
What is generated: the creature and location artwork, the interface, a large share of the code and the writing. What is measured rather than assumed: card balance, computed over thousands of simulated matches before every update. What is verified: the historical sources behind every legend, published one by one on the official site, where you can open them and prove me wrong.
What is never generated is your match. No image, no text and no game decision comes out of a model while you play: the engine is deterministic, and the solo opponent is a classic program applying the rules.
Crie au loup is one of the projects I run around generative AI, and the only one you can play. The making-of is documented without varnish in the devlog on the official site.
Some of this game's 2D art assets — creature, machine, prop and building sprites — were produced with the assistance of generative AI image tools during development. All generated material was reviewed, edited and hand-finished by our artists before it shipped, and it is integrated as fixed, pre-rendered art. The game does not generate any content live while you play.
Hostile Boards uses AI-generated content minimally, primarily for image creation and touch-ups in some of our visual assets. All AI-generated images are based on original designs and concepts created by our team. No live AI content generation or dynamic AI use occurs during gameplay
Generative AI was used during development to assist with portions of pre-generated in-game artwork and Steam store/marketing artwork. All outputs were reviewed, cleaned, composited, color-adjusted, and quality-checked by the development team. The game does not generate AI content at runtime.
Some graphic assets (club crests, flags, and interface icons) were created
with the help of generative AI tools during development. All content is
original and fictional, and does not reproduce any real brands, crests, or
logos. The match simulation, the game data, and the game code do not use AI
at runtime.
Some character artwork was developed from AI-assisted concept images, then
rebuilt and refined by the developer through 3D modelling and a deterministic
No generative AI is used while the game is running.
OpenAI ImageGen was used during development to generate and edit some pre-generated 2D pixel-art visual assets, including UI elements, icons, backgrounds, environment artwork, and character or animation source frames. All outputs were reviewed, selected, manually edited, cleaned up, resized, palette-normalized, and integrated by the developer as fixed game assets. The game does not use generative AI at runtime and does not generate live AI content.
As part of our early access work, we have used AI-generated images as placeholders for the illustrations on our cards and leader icons. This allows us to iterate quickly and see which game mechanics players like or dislike. We will replace these with human art once we've received enough feedback to finalize what we keep or discard.
Some images have been generated by AI. This include the in-game background as well as the Logo and other visual assets on the steam store associated with this game.
In ALASH, we utilized generative AI as a supplementary tool to assist in creating 2D art assets. Specifically, AI networks were used to help develop certain environment elements, textures, and background images to enhance the unique visual style of our alternate history setting. Please note that all game design, core mechanics, storyline, tower defense logic, and programming code were created entirely by hand by our development team. AI was used solely to support our artistic vision, ensuring that the integrity and soul of the gameplay remain entirely human-made.
Generative AI tools were used to support the creation and iteration of some visual assets and development-related content. All final assets are reviewed, edited, and selected by the developer before being included in the game or promotional materials. The game does not generate AI content live during gameplay.
Mad Friday is made by one person. The art, animation, design and code are all mine, made by hand. A few background music tracks are licensed royalty-free assets produced with a generative music model. All audio is pre-generated and shipped with the build; nothing is generated at runtime.
Localization into some supported languages was created with a mix of Community members and machine generation and reviewed by the developer.
The developer is actively seeking to replace any machine-generated localization over time.
During the development of this game, generative AI tools were used to assist in creating some of the art materials (scenes / props / UI) and text content. All AI-generated content has been thoroughly reviewed, modified, optimized, and checked for copyright by humans to ensure the quality and compliance of the content.
Generative AI tools were used to assist in the creation and refinement of some visual assets, including illustrations, user interface elements, and marketing materials.
Some visual assets in this game — including character portraits, illustrations, and marketing images — were created with the assistance of generative AI tools. All AI-generated material was reviewed, edited, and finalized by the developer. No AI content is generated during gameplay.
Some visual artwork and marketing assets were created or edited with generative AI under developer direction. All gameplay systems are authored, and no content is generated during gameplay.
Generative AI was used to assist in creating select background artwork and visual assets for the game. All assets were manually edited and refined before being integrated.
Tumble's visual art (character and charm illustrations, backdrops, item sprites, and store capsule art) was generated with Google's Gemini image models and then selected, cropped, retouched, colour-matched and in many cases regenerated by hand against a fixed reference style until it fit the game. Music and sound effects were generated with Suno under a commercial licence, then cut, mixed and placed by hand.
The game's design, rules, balance, and code are not generated content. Every mechanic, number and line of player-facing text was written, tested and revised by the developer, including through automated balance simulation and four rounds of human playtesting.
No content is generated live during play. Nothing in Tumble is produced at runtime by an AI model; the game runs entirely offline with no model calls of any kind.
AI has been used solely as a coding assistant for assistance purposes.
As a solo indie developer with no software background, using a coding assistant is my only option.
Regards.
We used Artificial Intelligence tools as an aid during development and in the creation of some game content. All AI-assisted material was personally reviewed and adjusted by our team, who were involved throughout every stage of the creative process.
本作在开发阶段使用了生成式 AI 辅助制作以下内容：美术（场景与背景、卡面插画、角色头像、界面位图、物品图标等游戏与商店页中的图像素材）由 AI 生成母图、经开发者逐项筛选、修复、裁切与组装后定稿；游戏内文本（事件与叙事文案）由开发者设计框架、AI 辅助起草、逐条经人工审查修改后定稿。游戏运行过程中不会实时生成任何 AI 内容，也不会连接任何 AI 服务。核心玩法、世界设定、事件结构与全部最终内容取舍均由开发者完成。
Final Night Firefight uses generative AI tools during development to assist with the creation of some pre-generated game artwork, visual assets and music. I review, edit, and integrate all AI-assisted content as part of the game's development process. The game does not generate AI content live during gameplay.
Generative AI was used to create the game's English and Simplified Chinese localization. AI tools were also used to assist with code generation during development. No AI-generated content is created during gameplay.
Some visual and audio assets in this game were created with the assistance of generative AI tools, and subsequently reviewed, edited, and refined by the developer. The game does not feature live or runtime generative AI during gameplay and does not generate content based on player input.
AI tools were used during the creation of draft visual assets. All AI-generated material is reviewed, edited, and manually refined by professional artists before being included in the final product.
Generative AI tools were used in the creation of certain assets for Camedieo.
Each was reviewed, reworked and integrated by us so the finished product holds a single consistent direction.
Generative AI tools were used during development to assist with the creation and iteration of some some pre-generated visual, written and promotional assets. All AI-assisted content is reviewed, edited and curated by the developer. No generative AI is used during gameplay.
Generative AI was used during development to assist in creating some 2D art assets, including scene backgrounds, character and monster artwork, UI elements, and icons for items, equipment, skills, and status effects. Generative AI was also used to assist in drafting some in-game text. All AI-assisted content was selected, edited, post-processed, and integrated by the developer. The game does not use generative AI at runtime and does not generate content in real time in response to player input.
Pocket Pawns was made with the help of generative AI, and the game's own credits name every tool used.
Images and music: the chess piece art was generated with ChatGPT, the board art with Gemini, and the world and environment art from assets created with Nano Banana Pro alongside free-to-use CC0 packs found online. The soundtrack is from Suno.
Claude wrote the game's code, its writing, and its translations into ten languages — and made much of what you see and hear directly in code rather than from an image generator: the character pixel art, the in-world props and interface, the three-dimensional rooms, the animation, and the sound effects.
The chess itself is not generated. The puzzles come from the Lichess puzzle database, released under CC0, and your opponent is Stockfish, a classical open-source chess engine. Real positions, a real engine.
Nothing is generated while you play. Pocket Pawns creates no content at runtime and never connects to an AI service. It does not need an internet connection to play.
Crie au loup is made with AI, from the first prototype to the illustrations. That is not a manufacturing detail: it is as much the project as the game itself.
My role is not a craftsman's: I painted no card, and I did not write the code line by line. I designed the system, I direct it, I arbitrate and I verify. Deciding what makes it into the game, what gets redone, what gets cut: that is production work, and it is mine.
What is generated: the creature and location artwork, the interface, a large share of the code and the writing. What is measured rather than assumed: card balance, computed over thousands of simulated matches before every update. What is verified: the historical sources behind every legend, published one by one on the official site, where you can open them and prove me wrong.
What is never generated is your match. No image, no text and no game decision comes out of a model while you play: the engine is deterministic, and the solo opponent is a classic program applying the rules.
Crie au loup is one of the projects I run around generative AI, and the only one you can play. The making-of is documented without varnish in the devlog on the official site.
Some of this game's 2D art assets — creature, machine, prop and building sprites — were produced with the assistance of generative AI image tools during development. All generated material was reviewed, edited and hand-finished by our artists before it shipped, and it is integrated as fixed, pre-rendered art. The game does not generate any content live while you play.
Hostile Boards uses AI-generated content minimally, primarily for image creation and touch-ups in some of our visual assets. All AI-generated images are based on original designs and concepts created by our team. No live AI content generation or dynamic AI use occurs during gameplay
Generative AI was used during development to assist with portions of pre-generated in-game artwork and Steam store/marketing artwork. All outputs were reviewed, cleaned, composited, color-adjusted, and quality-checked by the development team. The game does not generate AI content at runtime.
Some graphic assets (club crests, flags, and interface icons) were created
with the help of generative AI tools during development. All content is
original and fictional, and does not reproduce any real brands, crests, or
logos. The match simulation, the game data, and the game code do not use AI
at runtime.
Some character artwork was developed from AI-assisted concept images, then
rebuilt and refined by the developer through 3D modelling and a deterministic
No generative AI is used while the game is running.
OpenAI ImageGen was used during development to generate and edit some pre-generated 2D pixel-art visual assets, including UI elements, icons, backgrounds, environment artwork, and character or animation source frames. All outputs were reviewed, selected, manually edited, cleaned up, resized, palette-normalized, and integrated by the developer as fixed game assets. The game does not use generative AI at runtime and does not generate live AI content.
As part of our early access work, we have used AI-generated images as placeholders for the illustrations on our cards and leader icons. This allows us to iterate quickly and see which game mechanics players like or dislike. We will replace these with human art once we've received enough feedback to finalize what we keep or discard.
Some images have been generated by AI. This include the in-game background as well as the Logo and other visual assets on the steam store associated with this game.
In ALASH, we utilized generative AI as a supplementary tool to assist in creating 2D art assets. Specifically, AI networks were used to help develop certain environment elements, textures, and background images to enhance the unique visual style of our alternate history setting. Please note that all game design, core mechanics, storyline, tower defense logic, and programming code were created entirely by hand by our development team. AI was used solely to support our artistic vision, ensuring that the integrity and soul of the gameplay remain entirely human-made.
Generative AI tools were used to support the creation and iteration of some visual assets and development-related content. All final assets are reviewed, edited, and selected by the developer before being included in the game or promotional materials. The game does not generate AI content live during gameplay.
Mad Friday is made by one person. The art, animation, design and code are all mine, made by hand. A few background music tracks are licensed royalty-free assets produced with a generative music model. All audio is pre-generated and shipped with the build; nothing is generated at runtime.
Localization into some supported languages was created with a mix of Community members and machine generation and reviewed by the developer.
The developer is actively seeking to replace any machine-generated localization over time.
During the development of this game, generative AI tools were used to assist in creating some of the art materials (scenes / props / UI) and text content. All AI-generated content has been thoroughly reviewed, modified, optimized, and checked for copyright by humans to ensure the quality and compliance of the content.
Generative AI tools were used to assist in the creation and refinement of some visual assets, including illustrations, user interface elements, and marketing materials.
Some visual assets in this game — including character portraits, illustrations, and marketing images — were created with the assistance of generative AI tools. All AI-generated material was reviewed, edited, and finalized by the developer. No AI content is generated during gameplay.
Some visual artwork and marketing assets were created or edited with generative AI under developer direction. All gameplay systems are authored, and no content is generated during gameplay.
Generative AI was used to assist in creating select background artwork and visual assets for the game. All assets were manually edited and refined before being integrated.
