Google has unveiled Playground, an experimental Google Labs platform that lets you create a video game by writing down what you want, without a single line of code. The result runs straight in the browser, on a phone or a computer, and you can keep it to yourself, send it to a friend with a link or publish it in a public gallery.
The service launched on 7 October, but only for adult users in the United States. For Italian readers, then, it is for now a story to watch rather than a tool to try.
How it works
It all starts with a chat. You can begin from a blank page, pick up one of the starter suggestions or ask for help shaping a still vague idea. According to 9to5Google, you also choose at the outset whether the game will be 2D or 3D and whether it will be for one player or several.
Once a first version has been generated, you can try it straight away and keep refining it in plain words: physics, rules, characters and settings are changed simply by asking. According to 9to5Google, Google’s Gemini, Nano Banana and Lyria models sit behind it.
Sharing and the gallery
Games can be kept private, shared with a link or sent to Playground Explore, the public gallery. To highlight titles, the gallery relies on player votes and play activity. Some genres have leaderboards and multiplayer modes, and a Play Games profile lets you pick a name, save favourites and follow other creators.
Google says published games go through a safety check in line with its community guidelines and that anyone can report a piece of content. That is not a point to underestimate: a gallery of user-generated games, open to everyone, needs moderation from day one.
The next step: Unity Spark
The company has also previewed an integration with Unity Spark, aimed at those who want more advanced mechanics, professional-grade 3D graphics and the Unity runtime. It is being tested and a closed beta is on the way. No dates or access conditions have been given.
Who can use it, and at what price
The information here is still partial:
- available in the United States, for people aged 18 or over, at playground.google;
- access to creation is tiered and is being extended gradually according to your Google AI subscription;
- according to 9to5Google, even those who don’t pay can generate games, while wider access is tied to Google One.
Google has not announced dedicated pricing, a date for Italy or Europe, or the supported languages. If you’re interested, it’s worth keeping an eye on the company’s blog.
What we think
Playground is not the first attempt to bring generative AI to games, but three things make it interesting: it starts in the browser, so there is nothing to install; it brings creation, publishing and a gallery together in one place; and it targets an audience that can’t code.
The limits are obvious. A game born from a prompt risks looking like a thousand others, and quality depends a great deal on how well you can describe what you have in mind. The real test will be whether the gallery produces anything worth playing, or just a mountain of forgotten prototypes. The arrival of Unity Spark suggests Google wants to reach people looking for more than a pastime too, but for now it remains a promise.
Then there is the matter of access being tied to subscriptions. Google is organising its AI services across several paid plans, and the limits on Gemini’s models have also changed recently: we covered it in Gemini changes its plans: from 9 October the free account gets Flash-Lite only. How free Playground will really be, we will only find out when it arrives outside the United States.
For an Italian reader the advice is simple: no rush. When the service becomes available here, it’s worth trying it with a small idea, such as a skill game with a few rules, to see what it can really do before investing time in it.
Written by the SpazioiTech newsroom with the help of AI tools, from facts verified across at least two independent outlets, and checked before publication. Spotted a mistake? Tell us.
Translated from the Italian original. Read in Italian
