AI Games Are Coming

TLDR:

1. We are large believers in AI video games and are investing. Come speak to us.
2. We are internet hosting an invitation-only, free AI Games Summit on the NFX workplace in San Francisco on March 8, 2026, the day earlier than GDC begins. It’s sponsored by Silicon Valley Bank and Latitude Games. If you’re a founding father of an AI video games firm, apply here for an invitation. 
3. We are all nonetheless ready for AI video games to actually take off. There are technical and founder mindset obstacles, however the alternative continues to be sitting there.
4. With AI, we must always broaden our ideas of what a sport may be.

Why don’t we’ve an enormous AI sport but?

There are some early hits, like Death by AI, or AI-NPC pushed video games on Steam however nothing really big. Latitude (AI Dungeon and successors) might be the closest factor we’ve needed to an AI‑native breakout thus far – they noticed the chance early.

Given the ability of AI, it’s shocking we haven’t seen extra huge video games. There are a number of elements that designate why:

Foundational Technologies Are Still Arriving

Let’s take a look at what occurred with the web know-how window that opened in July 1993 and in addition didn’t produce video games instantly. 

As proven above, many nice shopper firms got here out within the first 6 years.  But no video games. They didn’t arrive for 7.5 – 16 years after the primary browser was obtainable and modems have been prevalent. Here are a few of them that got here inside that window of time.

• Runescape, Jan 2001
• SecondLife, June 2003 
• Steam, Sept 2003
• World of Warcraft, Nov 2004
• Roblox, Sept 2006
• Zynga, July 2007
• Minecraft, May 2009
• League of Legends, Oct 2009

The motive it took so lengthy is that the browser alone wasn’t enough. These video games have been left ready for bandwidth, video processing chips, reminiscence and different underlying applied sciences. Further, creating the V1.0 of those video games took 2-4 years in lots of circumstances.

As a primary instance, in 1995, Philip Rosedale needed to construct SecondLife, however he calculated the bandwidth and processing speeds of PCs have been NOT able to supporting the imaginative and prescient. So as an alternative, he constructed FreeVue, a video codec that was one of many foundational applied sciences that will be wanted for SecondLife to work. He bought FreeVue in 1996 and it grew to become the premise for Real Networks. He was CTO at Real till 1999 when he calculated that the opposite foundational applied sciences have been getting shut sufficient to assist SecondLife. He based a brand new firm to construct it, and launched the primary working model in June 2003. 

AI video games are in an analogous spot, ready for foundational applied sciences to be in place.  

The AI video games groups have been ready – and can proceed to attend – for pace of inference, price of inference, high quality of fashions, reminiscence administration strategies, in addition to the standard of photos, video, voice, and music. 

As you may see within the diagram above, the cellular know-how window was totally different.  Angry Birds, Supercell and Playtika all launched inside 3 years of the cellular app retailer opening up. Why? Because the foundational applied sciences have been all in place instantly on these magical cellphones.

Technology Windows

If you take a look at AI video games via the know-how window lens, we achieve a transparent image of why we don’t have a serious breakout now, and why we might have one quickly. You want to take a look at when a know-how window opens and when it closes. In this case, shopper software program merchandise are a great proxy. 

Here’s an instructive shopper software program instance. On the again of the digital pictures know-how window opening in 2002, Friendster began its beta in Oct 2002, after which launched March 2003. LinkedIn launched May 2003. In July 2003 three extra social networks launched in a 24-hour interval together with Tickle, Tribe and MySpace. In the following 4 months, eight extra launched like Orkut, Hi5, Piczo and others, after which Facebook launched in Feb 2004, eleven months after Friendster.  

Point is, the window to start out a successful social community was roughly eleven months.  

Once the cell phone arrived, one other know-how window opened up. We acquired WhatsApp Feb 2009, Instagram Oct 2010, and Snap in July 2011. So that cellular shopper social community window was open for roughly thirty months. After that window closed, it didn’t matter how good different merchandise have been, or how gifted the groups have been, or how onerous they labored. They couldn’t get any traction. 

These home windows are open for a short while and you must hit the market earlier than the window shuts.  

Our sense is the window for AI-native-games has simply cracked open within the final yr.

Founder Mindsets

The second problem for why we don’t have really huge AI video games but: we’re merely ready for the very best founders to wrap their heads round what is feasible with AI. Founders psychological fashions are nonetheless adjusting. The language we use to conceptualize and speak to one another continues to be forming.  

If founders can broaden their idea of what a sport may very well be, they may very well be on the doorstep of a breakthrough as we speak, even given the primitive stage of the foundational tech. 

What will we imply by broadening the idea of video games? See the diagram beneath which reveals different shopper experiences that are standard with AI thus far.

Today, in early 2026, it may very well be fertile to consider the merchandise of Roblox, Minecraft and WattPad. Open as much as concepts round AI twins, AI companions and the broad makes use of persons are discovering for ChatGPT.

Explore the place AI voice is as we speak already. Where picture era is. Video is nearly there in small clips.  What sort of sluggish movement experiences captivate gamers, even when ready for inference or picture era?

What types of experiences would enable as we speak’s hallucinating LLM’s to be simply forgiven or higher but, a function? 

Why will we imagine in AI video games a lot?

The video games business is greater than the movie and music industries mixed. Games IS leisure. We deal with the most important piece, the sport builders and publishers. Within that, we deal with cellular video games which is the most important piece, after which AI-first video games.

Given AI goes to alter every thing about video games over the following ten years we see an enormous prize for founders.

The incumbent sport builders will, after all, come at this chance from the place they sit. They will add AI incrementally to their present profitable enterprise and decrease prices, enhance pace, and enhance sport depth. All priceless. 

But we expect the breakthroughs, the brand new Robloxs, new Riots, new Supercells, will come from AI Native considering. Where founders begin with AI and take into consideration what engages individuals and what takes individuals on a journey.

New interfaces, new types of play, new participant sorts, new community results, new engagement, and new viral paths. In reality, there will likely be new methods of working sport improvement studios.

We need to guess with founders on the left facet of this diagram, going proper:

Come talk about it with us. The window is vast open. There is a lot to be invented within the subsequent 36 months. These markets haven’t been this open for 10 years.

Come be a part of us on the AI Games Summit March 8th by applying here.