For anyone getting a 403, it's US only right now and fails rather ungracefully. I just tried out a few games, framerates where strange/choppy, even in a match 3 game, and the games I tried were quite bad overall. I'm sure this type of thing will get good very soon though
- This is google experiment, meaning you're sadly likely invest more time in the product / community than google will from this point forward. (That's just my "fool me once" voice talking.)
- There really won't be much advantage to building in their playground vs building directly with any of the frontier models.
- Interest and distribution will be challenging because the bias is towards building your own game vs playing the games others make.
Every time I see a vibe-coded game, I think to "I should try that prompt" not "I want to play that game". I think personalized apps, experiences, and games are amazing but are a different beast and won't have the viral impact some creators are hoping for.
I've been vibe coding a game and it's so far from a one shot at this stage. A week and a half solid of prompting with multiple agents. Haven't looked at a line of code, but been iterating, playing and refining constantly. In my mind there's no chance anyone else could get the same result without pointing an agent at my game and reverse engineering it.
Maybe we need yet another word for this kind of development. It's going to be way more common.
Side note: I feel like I could only get this far with Opus 5.5 level models and tooling (cloud agents). No chance I could get to this point without rolling my sleeves up and helping with the technical side with earlier models, and no chance I could get this far pretty much only using my phone.
I'm basically paying Anthropic for both the experience of building a game with a team of decent engineers (it's been exhilarating), and the game itself. It's been incredibly fun. I would never pay this money for something to one shot it for me. This current level has been such a sweet spot of pretty much every ask being nailed and it still feeling like it's my own creation. Something one shotting it for me would kill most of the fun and feeling of "I did this" ownership.
i generally agree, but i can imagine a world where a vibe coded game hits some certain "threshold" where it has a few properties:
- easily distributable, i.e there is a very clear way to play it fairly quickly
- it doesn't feel like slop even though it's vibe coded
- it's super fun
I predict that will be games, very very few, but there will be some that hit that inflection point likely through Twitch streamers picking it up. But I also predict that they will be a flash in the pan. Even fewer of the already few, the ones that will be here to stay will likely need to turn into an actual mini studio and take the development/customers seriously.
Most of them will be fun projects that die the day they get released simply because building a decent looking vibe coded game is commoditized. Maybe some fun twitter engagements, but nothing more.
This is by far the hard part. The complete rulesets for most games[0] would be smaller than a prompt necessary to turn that board game into a computer game[1].
Anyone can draw some counters and a board, make up some rules. Making those rules interesting, let alone fun, is so much of a problem that many published board games get house-ruled in various ways[2].
[0] not all: I was obsessed with Star Fleet Battles as a young teenager, these days I describe it as "A computer game written about ten years before home computers had enough memory to store all the rules", which is why each turn takes 2 hours, one of which is counting out boxes and using that to fill out a spreadsheet.
[1] Unless the prompt was "here is a photo of the rules, turn it into a video game", but I count that as cheating, "a picture is worth a thousand words" etc.
[2] My main house rule for Star Fleet Battles, these days, is to simplify the 4+ kinds of power source into one. I can appreciate the world building of "an impulse engine can't divert antimatter to a photon torpedo", but it's not actually fun to keep track of this.
I don't see the appeal of scrolling through a list of games I could potentially make myself VS scrolling through a list of games I could immediately play.
We've hosted an AI game development platform for a while now and found there's a big difference between "making a game" and "making a fun game". I'd love to see if they are solving the real problem, but the link to try it out is a 403 for me.
I don't think AI will ever really solve "making a fun game" because:
1. Games are highly complex systems, so for example adding a fun mechanic from one game to another game can actually make that game less fun, and AI has no way to know, you just have to playtest it.
2. Games are evergreen (people still play and love the original Zelda/Mario/Tetris) and social (you want to share your love of game XYZ with others), so there's literally no reason for anyone to play a purely derivative game, when you can just play the original.
Eventually AI will be able to make perfect assets and all the other stuff AROUND making a fun game, but the game design will always be a human thing. At best, AI will come up with a bunch of game design experiments that a human will need to playtest and provide feedback on.
You may be right, at the same time a lot of "AI will never solve X" takes have aged pretty poorly, and I'm not confident what's fun about existing games cannot be codified and shown statistically somehow. Some genres seem more accessible to AI than others
I’m having a lot of fun creating an “improved” clone of a very old school game I loved.
It takes lots of back and forth but there is definitely (for me at least) to play this derivative game since it adds a lot of levels and stuff that weren’t available in the original.
Without AI I wouldn’t be able to do anything; but with it it can make some changes, I test them and we iterate on it.
So for me AI solves “making a fun game”.
It doesn’t “one shot GTA7” if that is your goal but that also sounds boring
Yea not the "making" part even (at least not yet). Every single demo/sample of a vibecoded* game thing is a piss poor "counter strike clone" or "gta clone" type of app that has bare bones everything (and i mean PS2 level or worse) and plays like absolute ass down to the basics of character control ans basic gameplay mechanics.
*: I don't consider a project you built with an LLM to be "vibecoded" if all the important ideas are from your head, you know what the code is supposed to be doing (domain model is in your head) and you guided the project every step of the way. Otherwise we're all vibecode. So yea, I don't doubt you can create amazing games with AI, but "with" as in "together", not as in "Claude, build me a Minecraft clone, make it fun, no mistakes."
this is probably a personal problem, but sometimes I get turned around on 'fun making a game' vs 'making a fun game' and I have the idea that ai elides this distinction even further
Beware: "you are on the list" - "or upgrade to access instantly"
I upgraded, same message. I downgraded, they kept the money because now I have a month of a plan that I didn't need to check something I still don't have access to.
A few of my friends and I have been building a game for a few weeks. It has been really fun because anyone can make a level and it’s somewhat random what level you get sent to- each level could have wildly different controls/goals, but getting oriented is part of the fun. We also have a ton of inside jokes and memories cooked in - I don’t think it would be very fun for someone external, it’s fun because a group of friends are building something together and that process is very fun. This seems to be just more of a game printing machine
This is why no code / low code is exciting. It will allow people who excel in what you call the "most important" parts to be able to create games.
Now, I don't think Google Playground is the best demonstration of this point, given it's intended to be quick turnaround. None of what's shown in the demo video is particularly thoughtful. But it still is a good indication of things to come.
From their blog[1]: "Playground launches today for users in the U.S. (18+) at playground.google, with tiered creation access rolling out based on your Google One membership."
How do we incorporate advertising/monetization that compensates us the creators into this?
Google Playground is a destination, currently, with no monetization. So any gamedev effort and ideas we contribute into it are collected by Google.
EDIT:
It is offered as "for funsies with your pals in America, bros!" and it looks a lot like it's almost a feature in the Google Play portfolio which makes me think this wasn't thought through fully.
So Google doing something strategic here which is not clear to me.
- are they training their AI?
- are they collecting user feedback for an eventual commercial launch?
If it's a great idea and the execution is fine, why not?
Eg Angry Birds. There are real world examples I have.
Why not?
edit: And about "your game" rhetoric, I'm talking about us more than me, unless you think that I specifically, NCR100, cannot create a game that is worthy? Why?
Where's the hope? Is there a presumption that AI generation is bad, full-stop?
Do you specifically assert that everything done with AI has no compelling quality to it?
I was taught that the tools do not make the artist.
When I sit down to build a product, a website, an app, middleware, servers, embedded apps, I expect to be paid if it is for a company. I will use other people's designs and patterns in my effort. I will work with everybody at the organization, even though who have, 'ideas'. It's my literal job as a Creator to make that slop compelling.
Why is the involvement of AI during creation a guarantor of non-compelling?
Google allow state-sponsored violent organizations but not non-state sponsored violence:
> Violent Organizations and Movements
> Known violent non-state organizations and movements are not permitted to use this product for any purpose. Do not create, upload, or distribute content that facilitates or promotes the activities of these groups, such as recruiting, coordinating online or offline activities, sharing manuals or other materials that could facilitate harm, promoting ideologies of violent non-state organizations, promoting terrorist acts, inciting violence, or celebrating attacks by violent non-state organizations. Depending upon the content, we may also take action against the user. Content related to violent non-state organizations may be allowed in an educational, documentary, scientific, or artistic context, but please be mindful to provide enough information to help people understand the context.
Maybe we need yet another word for this kind of development. It's going to be way more common.
Side note: I feel like I could only get this far with Opus 5.5 level models and tooling (cloud agents). No chance I could get to this point without rolling my sleeves up and helping with the technical side with earlier models, and no chance I could get this far pretty much only using my phone.
I'm basically paying Anthropic for both the experience of building a game with a team of decent engineers (it's been exhilarating), and the game itself. It's been incredibly fun. I would never pay this money for something to one shot it for me. This current level has been such a sweet spot of pretty much every ask being nailed and it still feeling like it's my own creation. Something one shotting it for me would kill most of the fun and feeling of "I did this" ownership.
- easily distributable, i.e there is a very clear way to play it fairly quickly
- it doesn't feel like slop even though it's vibe coded
- it's super fun
I predict that will be games, very very few, but there will be some that hit that inflection point likely through Twitch streamers picking it up. But I also predict that they will be a flash in the pan. Even fewer of the already few, the ones that will be here to stay will likely need to turn into an actual mini studio and take the development/customers seriously.
Most of them will be fun projects that die the day they get released simply because building a decent looking vibe coded game is commoditized. Maybe some fun twitter engagements, but nothing more.
This is by far the hard part. The complete rulesets for most games[0] would be smaller than a prompt necessary to turn that board game into a computer game[1].
Anyone can draw some counters and a board, make up some rules. Making those rules interesting, let alone fun, is so much of a problem that many published board games get house-ruled in various ways[2].
[0] not all: I was obsessed with Star Fleet Battles as a young teenager, these days I describe it as "A computer game written about ten years before home computers had enough memory to store all the rules", which is why each turn takes 2 hours, one of which is counting out boxes and using that to fill out a spreadsheet.
[1] Unless the prompt was "here is a photo of the rules, turn it into a video game", but I count that as cheating, "a picture is worth a thousand words" etc.
[2] My main house rule for Star Fleet Battles, these days, is to simplify the 4+ kinds of power source into one. I can appreciate the world building of "an impulse engine can't divert antimatter to a photon torpedo", but it's not actually fun to keep track of this.
Well ok then
1. Games are highly complex systems, so for example adding a fun mechanic from one game to another game can actually make that game less fun, and AI has no way to know, you just have to playtest it.
2. Games are evergreen (people still play and love the original Zelda/Mario/Tetris) and social (you want to share your love of game XYZ with others), so there's literally no reason for anyone to play a purely derivative game, when you can just play the original.
Eventually AI will be able to make perfect assets and all the other stuff AROUND making a fun game, but the game design will always be a human thing. At best, AI will come up with a bunch of game design experiments that a human will need to playtest and provide feedback on.
I’m having a lot of fun creating an “improved” clone of a very old school game I loved.
It takes lots of back and forth but there is definitely (for me at least) to play this derivative game since it adds a lot of levels and stuff that weren’t available in the original.
Without AI I wouldn’t be able to do anything; but with it it can make some changes, I test them and we iterate on it.
So for me AI solves “making a fun game”. It doesn’t “one shot GTA7” if that is your goal but that also sounds boring
*: I don't consider a project you built with an LLM to be "vibecoded" if all the important ideas are from your head, you know what the code is supposed to be doing (domain model is in your head) and you guided the project every step of the way. Otherwise we're all vibecode. So yea, I don't doubt you can create amazing games with AI, but "with" as in "together", not as in "Claude, build me a Minecraft clone, make it fun, no mistakes."
I upgraded, same message. I downgraded, they kept the money because now I have a month of a plan that I didn't need to check something I still don't have access to.
Now, I don't think Google Playground is the best demonstration of this point, given it's intended to be quick turnaround. None of what's shown in the demo video is particularly thoughtful. But it still is a good indication of things to come.
Happens when I click Press Start. Did Google kill this before they shipped it?
[1]https://blog.google/innovation-and-ai/technology/ai/playgrou...
oh playground... you were so young :'(
Google Playground is a destination, currently, with no monetization. So any gamedev effort and ideas we contribute into it are collected by Google.
EDIT:
It is offered as "for funsies with your pals in America, bros!" and it looks a lot like it's almost a feature in the Google Play portfolio which makes me think this wasn't thought through fully.
So Google doing something strategic here which is not clear to me.
- are they training their AI?
- are they collecting user feedback for an eventual commercial launch?
- are they building hype?
Why did Google spend the money on this?
If it's a great idea and the execution is fine, why not?
Eg Angry Birds. There are real world examples I have.
Why not?
edit: And about "your game" rhetoric, I'm talking about us more than me, unless you think that I specifically, NCR100, cannot create a game that is worthy? Why?
Where's the hope? Is there a presumption that AI generation is bad, full-stop?
Do you specifically assert that everything done with AI has no compelling quality to it?
I was taught that the tools do not make the artist.
When I sit down to build a product, a website, an app, middleware, servers, embedded apps, I expect to be paid if it is for a company. I will use other people's designs and patterns in my effort. I will work with everybody at the organization, even though who have, 'ideas'. It's my literal job as a Creator to make that slop compelling.
Why is the involvement of AI during creation a guarantor of non-compelling?
> Violent Organizations and Movements
> Known violent non-state organizations and movements are not permitted to use this product for any purpose. Do not create, upload, or distribute content that facilitates or promotes the activities of these groups, such as recruiting, coordinating online or offline activities, sharing manuals or other materials that could facilitate harm, promoting ideologies of violent non-state organizations, promoting terrorist acts, inciting violence, or celebrating attacks by violent non-state organizations. Depending upon the content, we may also take action against the user. Content related to violent non-state organizations may be allowed in an educational, documentary, scientific, or artistic context, but please be mindful to provide enough information to help people understand the context.