--- title: Think You Can Build a Game with AI? Think Again! -... description: Two Meta engineers share their experience leading AI-driven game creation work over the past year. They demonstrate how AI tools are lowering barriers for both... canonical: https://daily.dev/posts/think-you-can-build-a-game-with-ai-think-again---danielle-an-david-hoe-meta-n2ijabgs3 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Think You Can Build a Game with AI? Think Again! - Danielle An & David Hoe, Meta | daily.dev og:description: Two Meta engineers share their experience leading AI-driven game creation work over the past year. They demonstrate how AI tools are lowering barriers for both... og:url: https://daily.dev/posts/think-you-can-build-a-game-with-ai-think-again---danielle-an-david-hoe-meta-n2ijabgs3 og:image: https://api.daily.dev/og/posts/N2IJAbgS3.png og:image:alt: Think You Can Build a Game with AI? Think Again! - Danielle An & David Hoe, Meta og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Think You Can Build a Game with AI? Think Again! - Danielle An & David Hoe, Meta **[AI Engineer](https://daily.dev/sources/aidotengineer)** · 18 min read · 0 upvotes · 0 comments ## Summary Two Meta engineers share their experience leading AI-driven game creation work over the past year. They demonstrate how AI tools are lowering barriers for both coders and artists to build games, while noting that aesthetics, cohesion, and taste still differentiate good games from bad ones. Key highlights include using key art as an anchor for generative models to maintain visual consistency, and runtime LLMs powering non-scripted NPC behavior in multiplayer games. They also discuss how AI enables parallel team workflows instead of linear waterfall processes, and flag open engineering challenges around non-determinism, debugging agentic systems at scale, and content safety when LLMs generate in-game content at runtime. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is a runtime LLM in game development and how does it change gameplay? A runtime LLM is a living AI entity that operates while a game is being played, making independent decisions that modify or direct the game in real time rather than following scripted behavior. In a multiplayer demo built by a Meta team in a couple of days, LLM-driven NPCs given personalities like thief or honorable independently chose to steal cubes, block, or kick other NPCs, making every playthrough unique and non-repeatable. _Developers experimenting with runtime LLM-driven NPCs can follow emerging techniques and demos on daily.dev._ ### How has AI changed the traditional game development pipeline? AI has shifted game development away from a linear waterfall process, where design, art, modeling, animation, and coding happened sequentially and revisiting upstream decisions was costly, toward teams working in parallel with iterations completed in hours or days instead of months. This allows more playtesting and more time to refine game ideas rather than executing a fixed plan linearly. _Teams rethinking their production pipeline around AI tooling can track workflow shifts like this on daily.dev._ ### What challenges remain for using AI and agentic systems in game platforms at scale? Key open challenges include maintaining stability and determinism when models and prompts change unpredictably, debugging agentic systems across the stack from user prompting to runtime LLM decisions to content ranking and delivery, managing token economics so creators, players, and platforms stay profitable, and ensuring safety of dynamically generated content for audiences. _Engineers tackling agentic system reliability and AI content safety can find related discussions on daily.dev._ ## Similar posts on daily.dev - [The next great game may not come from a million-dollar studio](https://daily.dev/posts/the-next-great-game-may-not-come-from-a-million-dollar-studio-gkujpi1yc) · The Next Web · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#game-development](https://daily.dev/tags/game-development) [View this post on daily.dev](https://daily.dev/posts/think-you-can-build-a-game-with-ai-think-again---danielle-an-david-hoe-meta-n2ijabgs3) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Think You Can Build a Game with AI? Think Again! - Danielle An & David Hoe, Meta","url":"https://daily.dev/posts/think-you-can-build-a-game-with-ai-think-again---danielle-an-david-hoe-meta-n2ijabgs3","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/think-you-can-build-a-game-with-ai-think-again---danielle-an-david-hoe-meta-n2ijabgs3"},"datePublished":"2026-07-08T15:19:30.209Z","dateModified":"2026-09-13T19:02:12.728Z","description":"Two Meta engineers share their experience leading AI-driven game creation work over the past year. They demonstrate how AI tools are lowering barriers for both...","image":"https://i.ytimg.com/vi/grdoOC1BT1s/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/grdoOC1BT1s/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/think-you-can-build-a-game-with-ai-think-again---danielle-an-david-hoe-meta-n2ijabgs3","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,game-development","timeRequired":"PT18M","video":{"@type":"VideoObject","name":"Think You Can Build a Game with AI? Think Again! - Danielle An & David Hoe, Meta","description":"Two Meta engineers share their experience leading AI-driven game creation work over the past year. They demonstrate how AI tools are lowering barriers for both...","thumbnailUrl":"https://i.ytimg.com/vi/grdoOC1BT1s/sddefault.jpg","uploadDate":"2026-07-08T15:19:30.209Z","duration":"PT18M","url":"https://api.daily.dev/r/N2IJAbgS3","embedUrl":"https://www.youtube.com/embed/grdoOC1BT1s"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"Think You Can Build a Game with AI? Think Again! - Danielle An & David Hoe, Meta"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/think-you-can-build-a-game-with-ai-think-again---danielle-an-david-hoe-meta-n2ijabgs3#faq","mainEntity":[{"@type":"Question","name":"What is a runtime LLM in game development and how does it change gameplay?","acceptedAnswer":{"@type":"Answer","text":"A runtime LLM is a living AI entity that operates while a game is being played, making independent decisions that modify or direct the game in real time rather than following scripted behavior. In a multiplayer demo built by a Meta team in a couple of days, LLM-driven NPCs given personalities like thief or honorable independently chose to steal cubes, block, or kick other NPCs, making every playthrough unique and non-repeatable. Developers experimenting with runtime LLM-driven NPCs can follow emerging techniques and demos on daily.dev."}},{"@type":"Question","name":"How has AI changed the traditional game development pipeline?","acceptedAnswer":{"@type":"Answer","text":"AI has shifted game development away from a linear waterfall process, where design, art, modeling, animation, and coding happened sequentially and revisiting upstream decisions was costly, toward teams working in parallel with iterations completed in hours or days instead of months. This allows more playtesting and more time to refine game ideas rather than executing a fixed plan linearly. Teams rethinking their production pipeline around AI tooling can track workflow shifts like this on daily.dev."}},{"@type":"Question","name":"What challenges remain for using AI and agentic systems in game platforms at scale?","acceptedAnswer":{"@type":"Answer","text":"Key open challenges include maintaining stability and determinism when models and prompts change unpredictably, debugging agentic systems across the stack from user prompting to runtime LLM decisions to content ranking and delivery, managing token economics so creators, players, and platforms stay profitable, and ensuring safety of dynamically generated content for audiences. Engineers tackling agentic system reliability and AI content safety can find related discussions on daily.dev."}}]} ```