Public Steam disclosure research25,380 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.
We leverage AI technology in certain tasks to achieve higher-quality results. AI is used primarily for handling repetitive tasks or as a tool for quality enhancement, while the core content and final outputs are crafted by skilled professionals."
AI was used in the process of creating the images used in the game. Each element generated by the neural network was carefully edited and refined by a 2D artist to ensure high quality content and a unified artistic style of in-game visual materials.
The store page of this game and some game materials have been polished by the open-source AI (stable diffusion), and further processed and optimized by the developers. All drafts are photos taken by the developers themselves and line drawings drawn by themselves. We promise that all AI-generated content complies with the requirements of relevant laws and regulations.
The Everest Encounter leverages cutting-edge AI to create a uniquely immersive VR experience. Our game features Katherine, an AI-driven flight attendant powered by advanced language models and NVIDIA's AI technology. Players engage in natural conversations and dynamic interactions, with Katherine adapting her responses and behavior based on the player's choices and emotions. While the game's core content is pre-rendered for optimal performance, Katherine's dialogue and decision-making occur in real-time, ensuring each playthrough is unique. This innovative use of AI pushes the boundaries of narrative gaming, offering players an unparalleled level of engagement and realism in a virtual world.
The game uses images generated by AI to illustrate the story and offer visual interactions.
The main content of game, the text of the story, is NOT using AI.
We used AI for some detail optimization in our artwork, but we guarantee that the majority of the art assets are handcrafted.
We will never use 100% AI-generated artwork for profit. Everything you see has been manually redrawn at least once.
AI-generated materials are not used directly.
of some of the character illustrations are used
Posing, facial expressions, and composition of some character illustrations are used as references.
I draw by hand while looking at the AI-generated materials.
NOTES: This work was set up in September 2022. For the purpose of conducting academic research on emerging technologies, AI image generation technology (which was still in its early stages of development at the time) was used to assist in processing the visual effects of some content, and an AI music generation tool was used to create the background music. In order to avoid the abuse of artificial intelligence generation tools, only the default data sets provided by AI tools are used during the development process. All usage methods include:
1. Process the real photos to create "alien" and "distorted" effects;
2. With the "abstract image generated based on the provided keywords", manually perform secondary design, reconstruction, and drawing;
3. Generate music clips that fit the atmosphere by adjusting parameters such as musical instruments used and pace, then perform secondary editing to them.
All contents involving artificial intelligence were completed before 2023. After that, AI image generation technology has been completely stopped, no longer and will no longer continue to be used for development.
For all AI-generated image assets utilized in the work, this work only reserves the right to use.
Some in-game or store assets might be produced through processes involving the use of AI filters or AI generated elements merged with other sources. Whenever using a generative AI, we exclusively use private models trained with licensed sets of data.
All of the images in this game are generated by neural networks trained on images (aka “AI generated”). If you choose to boycott all AI generated material, please do not buy this game. If you are uncomfortable with the thought but are interested in my game, please read on.
The aesthetic of this game had to come from the strange, uncanny look of AI. There is a heartlessness to the visuals that is intentional. I am not trying to copy the look of other artists. The only tool capable of producing the look I was going for was a malfunctioning AI. Ironically, I have had a number of people tell me that it looks much more realistic than standard AI. I am assuming this is because of how I tried to reproduce the look of old media.
All images, aside from the title screen, credits and some of the video shot on camcorder were generated by AI prompts. It took several days of exploring before I began generating images that fit the aesthetic I was looking for. I have heard AI described as a slot machine, and I think the metaphor is somewhat accurate. I think of this as similar to indeterminacy in music. Using random number generators can produce interesting outputs, but you need to refine and direct the system to do so.
I started my process in Midjourney. My ratio of usable images was around 100 to 1; for every 100 images I generated, I found one usable piece for the game. I created an automated process to transform these images to 16:9 widescreen in Photoshop Beta. Some of these images I touched up manually or regenerated sections of the image. There are some real images and hand painted imagery within the AI elements. Often, I would have to generate the sides of the image multiple times because photoshop didn’t generate what I needed.
To those concerned about not paying artists: I understand this, but I want to be clear that this was not created to replace the work of artists, any more than the synthesized scores to horror films were about replacing orchestras. These are tools that do different things. Rather than some sort of lazy shortcut, the use of AI imagery is essential to the aesthetic and narrative of the Path of Veins.
Many images that take place in game, representing towns and cities were created with the help of artificial intelligence. Generated pieces were not directly taken place in game but used advisory/recommendation manner (eg. to illustrate how a North West city looks in 1940s).
Some of the art materials for this work were created using Stable Diffusion, including but not limited to:
"CG" used in the work;
Partial small elements.
Note: The right to interpret the AI content of this work belongs to the person in charge, Liu, and any consequences of AI painting are borne by the first responsible person, Liu.
NO AI WAS USED IN THE CREATION OF ANY ART ASSETS, MUSIC OR WRITING. AI IS ONLY USED FOR CHARACTER INTERACTIONS AND MECHANICS.
Built with Llama
All AI models are run on the user's machine using their GPU. No data usage is sent to us and the game can be played offline.
The AI is character-based. Each character has their own model and will respond and react to player interaction in a specific way, each providing a unique challenge. As the player progresses through the year in Millennium Whisper, they will build relationships with the characters as all interactions are fed to the models and the characters rate their affection levels towards the player. The challenge comes from the player needing to juggle their social relationship with all the characters at once as characters have the ability to communicate with eachother by spreading rumours thereby allowing neglected characters to sabotage the player's relationship with other characters. The characters will also accept or reject date invites from the player or author their own date invitations. At the end of the game the AIs evaluates which character the player ends up with.
Only a small number of placeholder sound effects, assets, and promotional illustrations are used in the game. They will be gradually replaced with original handmade content not generated by AI. All core assets of the game are handmade.
In the development of our game, we incorporated AI tools to enhance various aspects of the creative process. Our game uses generative models for creating and editing images, sound mixing, and text translation. These AI tools were utilized to improve quality and efficiency, allowing the development team to concentrate on crafting the narrative, gameplay, and overall design. The AI played a supportive role, assisting with technical tasks while the core vision and creative direction were driven by the team.
For the development of The Broken Mirror (Xuru Engine Edition), we have integrated advanced generative AI workflows to enhance the atmospheric Noir experience.
Environment & Visual Arts: Environmental assets and backgrounds have been conceptualized and generated using AI, then manually refined and stylized in high-contrast black and white to achieve a unique 'graphic novel' aesthetic.
Audio Experience: Soundtrack and atmospheric sound effects were developed with AI assistance to create a deep, immersive psychological thriller mood.
Human Oversight: All AI-assisted content has been curated, edited, and directed by our lead developer. The core narrative, character arcs, complex puzzle design, and final assembly are 100% original human creations, ensuring a cohesive and high-quality artistic vision.
All main character illustrations and story CGs in the game are hand-drawn by artists.
Some background images use free AI-generated background assets licensed for commercial use.
In addition, AI was used as a supplementary reference during the creation of certain scene assets.
We do not use unedited AI-generated content - every image in the game is handcrafted by our 2D art team. However, they use AI-generated images as rough drafts that are later polished or fully reworked by the 2D art team. We only use AI-generated images as drafts for art assets. All other assets - texts, music, UI - are 100% human-made. We would not want to diminish the efforts of our 2D art team who spent countless hours working on the game to bring you the best experience - we just feel the need to be transparent. We also want to highlight that all images are processed by human artists in the end.
We leverage AI technology in certain tasks to achieve higher-quality results. AI is used primarily for handling repetitive tasks or as a tool for quality enhancement, while the core content and final outputs are crafted by skilled professionals."
AI was used in the process of creating the images used in the game. Each element generated by the neural network was carefully edited and refined by a 2D artist to ensure high quality content and a unified artistic style of in-game visual materials.
The store page of this game and some game materials have been polished by the open-source AI (stable diffusion), and further processed and optimized by the developers. All drafts are photos taken by the developers themselves and line drawings drawn by themselves. We promise that all AI-generated content complies with the requirements of relevant laws and regulations.
The Everest Encounter leverages cutting-edge AI to create a uniquely immersive VR experience. Our game features Katherine, an AI-driven flight attendant powered by advanced language models and NVIDIA's AI technology. Players engage in natural conversations and dynamic interactions, with Katherine adapting her responses and behavior based on the player's choices and emotions. While the game's core content is pre-rendered for optimal performance, Katherine's dialogue and decision-making occur in real-time, ensuring each playthrough is unique. This innovative use of AI pushes the boundaries of narrative gaming, offering players an unparalleled level of engagement and realism in a virtual world.
The game uses images generated by AI to illustrate the story and offer visual interactions.
The main content of game, the text of the story, is NOT using AI.
We used AI for some detail optimization in our artwork, but we guarantee that the majority of the art assets are handcrafted.
We will never use 100% AI-generated artwork for profit. Everything you see has been manually redrawn at least once.
AI-generated materials are not used directly.
of some of the character illustrations are used
Posing, facial expressions, and composition of some character illustrations are used as references.
I draw by hand while looking at the AI-generated materials.
NOTES: This work was set up in September 2022. For the purpose of conducting academic research on emerging technologies, AI image generation technology (which was still in its early stages of development at the time) was used to assist in processing the visual effects of some content, and an AI music generation tool was used to create the background music. In order to avoid the abuse of artificial intelligence generation tools, only the default data sets provided by AI tools are used during the development process. All usage methods include:
1. Process the real photos to create "alien" and "distorted" effects;
2. With the "abstract image generated based on the provided keywords", manually perform secondary design, reconstruction, and drawing;
3. Generate music clips that fit the atmosphere by adjusting parameters such as musical instruments used and pace, then perform secondary editing to them.
All contents involving artificial intelligence were completed before 2023. After that, AI image generation technology has been completely stopped, no longer and will no longer continue to be used for development.
For all AI-generated image assets utilized in the work, this work only reserves the right to use.
Some in-game or store assets might be produced through processes involving the use of AI filters or AI generated elements merged with other sources. Whenever using a generative AI, we exclusively use private models trained with licensed sets of data.
All of the images in this game are generated by neural networks trained on images (aka “AI generated”). If you choose to boycott all AI generated material, please do not buy this game. If you are uncomfortable with the thought but are interested in my game, please read on.
The aesthetic of this game had to come from the strange, uncanny look of AI. There is a heartlessness to the visuals that is intentional. I am not trying to copy the look of other artists. The only tool capable of producing the look I was going for was a malfunctioning AI. Ironically, I have had a number of people tell me that it looks much more realistic than standard AI. I am assuming this is because of how I tried to reproduce the look of old media.
All images, aside from the title screen, credits and some of the video shot on camcorder were generated by AI prompts. It took several days of exploring before I began generating images that fit the aesthetic I was looking for. I have heard AI described as a slot machine, and I think the metaphor is somewhat accurate. I think of this as similar to indeterminacy in music. Using random number generators can produce interesting outputs, but you need to refine and direct the system to do so.
I started my process in Midjourney. My ratio of usable images was around 100 to 1; for every 100 images I generated, I found one usable piece for the game. I created an automated process to transform these images to 16:9 widescreen in Photoshop Beta. Some of these images I touched up manually or regenerated sections of the image. There are some real images and hand painted imagery within the AI elements. Often, I would have to generate the sides of the image multiple times because photoshop didn’t generate what I needed.
To those concerned about not paying artists: I understand this, but I want to be clear that this was not created to replace the work of artists, any more than the synthesized scores to horror films were about replacing orchestras. These are tools that do different things. Rather than some sort of lazy shortcut, the use of AI imagery is essential to the aesthetic and narrative of the Path of Veins.
Many images that take place in game, representing towns and cities were created with the help of artificial intelligence. Generated pieces were not directly taken place in game but used advisory/recommendation manner (eg. to illustrate how a North West city looks in 1940s).
Some of the art materials for this work were created using Stable Diffusion, including but not limited to:
"CG" used in the work;
Partial small elements.
Note: The right to interpret the AI content of this work belongs to the person in charge, Liu, and any consequences of AI painting are borne by the first responsible person, Liu.
NO AI WAS USED IN THE CREATION OF ANY ART ASSETS, MUSIC OR WRITING. AI IS ONLY USED FOR CHARACTER INTERACTIONS AND MECHANICS.
Built with Llama
All AI models are run on the user's machine using their GPU. No data usage is sent to us and the game can be played offline.
The AI is character-based. Each character has their own model and will respond and react to player interaction in a specific way, each providing a unique challenge. As the player progresses through the year in Millennium Whisper, they will build relationships with the characters as all interactions are fed to the models and the characters rate their affection levels towards the player. The challenge comes from the player needing to juggle their social relationship with all the characters at once as characters have the ability to communicate with eachother by spreading rumours thereby allowing neglected characters to sabotage the player's relationship with other characters. The characters will also accept or reject date invites from the player or author their own date invitations. At the end of the game the AIs evaluates which character the player ends up with.
Only a small number of placeholder sound effects, assets, and promotional illustrations are used in the game. They will be gradually replaced with original handmade content not generated by AI. All core assets of the game are handmade.
In the development of our game, we incorporated AI tools to enhance various aspects of the creative process. Our game uses generative models for creating and editing images, sound mixing, and text translation. These AI tools were utilized to improve quality and efficiency, allowing the development team to concentrate on crafting the narrative, gameplay, and overall design. The AI played a supportive role, assisting with technical tasks while the core vision and creative direction were driven by the team.
For the development of The Broken Mirror (Xuru Engine Edition), we have integrated advanced generative AI workflows to enhance the atmospheric Noir experience.
Environment & Visual Arts: Environmental assets and backgrounds have been conceptualized and generated using AI, then manually refined and stylized in high-contrast black and white to achieve a unique 'graphic novel' aesthetic.
Audio Experience: Soundtrack and atmospheric sound effects were developed with AI assistance to create a deep, immersive psychological thriller mood.
Human Oversight: All AI-assisted content has been curated, edited, and directed by our lead developer. The core narrative, character arcs, complex puzzle design, and final assembly are 100% original human creations, ensuring a cohesive and high-quality artistic vision.
All main character illustrations and story CGs in the game are hand-drawn by artists.
Some background images use free AI-generated background assets licensed for commercial use.
In addition, AI was used as a supplementary reference during the creation of certain scene assets.
We do not use unedited AI-generated content - every image in the game is handcrafted by our 2D art team. However, they use AI-generated images as rough drafts that are later polished or fully reworked by the 2D art team. We only use AI-generated images as drafts for art assets. All other assets - texts, music, UI - are 100% human-made. We would not want to diminish the efforts of our 2D art team who spent countless hours working on the game to bring you the best experience - we just feel the need to be transparent. We also want to highlight that all images are processed by human artists in the end.