Find us!

X @AdaEden1001 | Tiktok @@1001nights.ai

Join our Discord through the right sidebar!

關於此遊戲

A game where the power of stories shapes reality, driven by AI language model...

Adapted from Persian folklore: The brutal King Shahryar, enraged by suspected betrayal of his wife, takes a new bride each day only to execute her at dawn. You are Shahrzad, his next bride...

Can you end his reign of terror through your stories?

Inspired by Wittgenstein's notion: "The limits of my language mean the limits of my world."

Unleash your imagination! Tell captivating tales to intrigue the AI-powered King, cleverly guiding him to mention words like 'sword', 'knife', or 'shield' - watch as they materialize into real weapons!

Brought to life by cutting-edge AI language models, each story is a unique creation - completely unscripted! But beware... strange words might provoke the King's wrath...

Collect weapons forged from your stories!

Generated in real-time based on your narratives! From shawarma kitchen knives to witch's spears... every weapon is one of a kind!

Battle the King in a world born from your tales!

Each battle is an unprecedented theatrical performance - weapon effects and even character dialogues spring from the stories you've crafted!

Features:

AI-Native Game:

• Powered by generative AI (especially large language models) as core gameplay mechanics. In other words, this innovative experience would be impossible without AI. Award-winning gameplay ensures every story is unique!

Pixel Art Aesthetics:

• Features distinctive pixel art and hand-drawn CGs! AI is used solely for the innovative storytelling gameplay - all artwork is created by human artists.

Challenge Traditional Norms:

• Break free from ancient patriarchal constraints with Shahrzad!

Academic Foundation:

• More than just a game - it's an academic and artistic project. We've published two papers on the game's systems, hoping to inspire more novel games!

AI 生成內容聲明

開發者針對他們的遊戲使用 AI 生成內容方式的說明如下：

As an AI-native game that started development as early as 2020, before the release of various AI language/image models, our goal is to explore creative narrative gameplay based on conversational AI to bring players novel experiences they've never had before: telling stories, and turning stories into reality.
All game art assets are created by human artists. Specifically, this refers to assets that can be pre-prepared and reused, including but not limited to character CGs, scene models, UI, etc. The dynamically generated content based on player inputs: [story text] and [story scene images], are generated by AI. This is because the game is based on the player’s free input and requires real-time interaction (e.g., the King continuing the story), which cannot be pre-prepared in any way.
When we started this game as an art and academic project, our guiding principle was to use AI only to create novel gaming experiences that were previously impossible, not to replace human creativity in any way. In other words, without AI technology, the game mechanics could not be realized. Inspired by the ancient Persian folklore, the game's soul comes from us, the human creators, not the AI technology itself. AI is just the means by which this vision is brought to life and shared with you. As human artists and researchers, we hope this game serves as an example of how technology can be an actor, while human creators become the directors who shape it in artistic and innovative ways.
The main game mechanics are as follows:
1. You narrate stories to the King, who is powered by a large language model (the LLM in our free trial is GLM, he continues your tale. You must guide him to mention specific weapon keywords. Each weapon is unique, with descriptions generated based on the stories you and the King create together.
2. As you collect more weapons, images of the world in the story will appear on the weapon cards and in the game scenes. This is part of the gameplay we designed, namely 'turning stories into reality.' This can only be generated in real-time based on player inputs (e.g., the player tells a story about a forest cat, and then sees a cat on the card), and cannot be pre-prepared in any way. Technically, we use open-source CheeseDaddy and revAnimated as the base, combined with a licensed Pixelization algorithm.
We use a stylistic LoRa model trained entirely on art created by our team. All training images come from the game’s CGs and artworks that our team members have authorized for use. This ensures that the generated images align only with our original "1001 Nights" art style, avoiding any resemblance to external artworks and ensuring data sources and generated content are secure and consistent.
The basics of above workflow has been published in our 2 academic papers. You can read it on the official game website on the right.
Currently, we're a small and self-funded indie team without any funding or income from the game. The game demo is 100% free and does not use player data for AI training. We are well aware that AI technology still has ethical controversies and challenges, and we understand players' concerns. It is our responsibility to listen to everyone's suggestions and feedback. With appropriate future funding and support, we aim to address and reduce potential problems from generative AI technologies. We welcome any concerns, suggestions, or questions about our AI-generated content and processes via adaeden1001[at]gmail[dot]com or through our Discord/QQ. Your feedback will help us improve and become more considerate.