Can AI Be Used to Solve Gaming’s Saturation Problem?

It’s fair to say that gamers do not lack for choice these days. The app stores host several hundred thousand games; Steam hosts over 130,000 as of today; console titles will number in the tens of thousands, and that’s before we add in online browser games on various platforms, not to mention classics reproduced for emulators. It is an abundance of choice. Obviously, you won’t get the chance to play them all, and most people – even the most avid gamers – will only ever be able to play a small fraction of the releases.

Unfortunately, the logic also dictates that a lot of good games remain undiscovered. That shows up in the data on platforms like Steam, where the majority of games struggle to make revenue of over a few hundred dollars; some make nothing at all. Of course, some of those games will be objectively terrible, but many that are excellent simply remain undiscovered and unprofitable. 

Games can be difficult to describe for search engines 

Gaming is very broad in terms of genres and styles, and so many individual elements can be difficult to compartmentalize for basic search engines. It’s different to, say, online casino platforms, where games have defined styles – roulette, card games, slots, etc. – and even movie and TV streaming sites like Netflix make it easy to get what you are looking for pinned down. 

Sites like Steam have most certainly tried to address the issue, and there are external tools, such as Ludocene, which is a game discovery engine, but perhaps AI could make things a little easier? Of course, you could always describe your gaming tastes to Gemini or ChatGPT and see what they recommend, but we can see a future where AI is built into the search results on platforms like Steam. 

A system where AI creates a Tinder for gaming 

Indeed, if we return to Ludocene – it’s not run by AI, but recommendations are based on expert gaming reporters – the idea behind that discovery engine is that it is like Tinder for gaming, basically trying to match you based on your profile. Perhaps you like games based on problem-solving, but you don’t really like classic puzzle games. Maybe you like FPS games, but do not like war games.

There are a lot of broad elements that don’t usually fit neatly into a search engine, so AI could perhaps parse out some of what you like and match to games that might suit your personal tastes. Indeed, much of AI’s future lies in its ability to deliver personal recommendations better than an algorithm can. It’s already being incorporated into broader search engines – look at what Google announced at I/0 2026 – so it makes sense that gaming platforms do the same. 

Will this solve the saturation problem? Probably not in its entirety. There are simply too many games, and, perhaps ironically, AI and vibe coding will probably lead to a situation where even more games flood the market. But the technology feels like it is key to helping with discoverability, allowing individuals to find their perfect gaming partner among the many, many undiscovered titles out there.