Meta Moves Game Creation From Code to Prompts With Horizon Tools
Meta has announced two tools that let people create games for its Horizon social platform with AI prompts. The Verge
Horizon Create is a mobile app for that prompt-based creation. Horizon Studio does the same job in a web browser, with more granular controls for creators who want finer adjustment.
Both tools are planned for early access. Interested users can sign up for a waitlist for Horizon Create and Horizon Studio.
Distribution is built in. Games published with either tool can be recommended on Facebook and Instagram and played there. That puts Horizon games inside feed-driven discovery and mobile play, rather than confining them to the Horizon client alone.
Meta says both tools can produce complete 2D or 3D mobile games with progression systems, balanced difficulty, art direction, and multiplayer. In plain terms, that means unlocks and levels, fairly tuned challenge, a consistent visual style, and online sessions with others, not only single scenes.
The broader context here is a change in the authoring interface. Instead of placing each object and writing code for how it behaves — a bit like laying every brick — creators describe the systems they want and the toolchain builds a first version. For working developers, the familiar questions apply: what is visible and editable after generation, and how consistent is the output when the prompt changes.
In my view, the technical test will be those four promised features. Progression needs saved state, unlock rules, and a balanced in-game economy. Balanced difficulty needs data from real play and tuning curves, not a single automatic pass. Art direction needs matching assets, lighting, and menus. Multiplayer needs synced game states, handling of network delay, and session management. Each is easy to demo and hard to sustain for live players over time.
Looking at what this means for creators, the split between phone and browser points to a tiered workflow, with fast tries on mobile and deeper revision in Studio. The practical issue is round-tripping. If a game starts from a prompt on a phone, can a team open that same project in a browser, inspect the generated logic, change parts by hand, and keep those changes when the prompt is revised.
Stepping back to consider distribution, the Facebook and Instagram detail matters as much as generation. Games that live in a feed need short sessions, fast loading, touch input, and visuals that are clear at a glance. A complete game in this context means one that holds up against competing posts, cold starts, and interrupted sessions.
In this author's view, the optimistic case is that lower authoring cost allows more experimentation. When progression, tuning, art, and multiplayer basics arrive by default, small teams can test game loops that would otherwise require substantial early engineering. We have seen this pattern before with higher-level game engines, with more prototypes, faster cuts of weak ideas, and more time left for design and live operation.
