This is part of 30 consecutive series of posts related to understanding and modeling cryptocurrency-related economic activity. We are sifting through hype and speculation to find quantitative insight.
In previous posts, I wrote,
Games like Roblox that sell in-game currency don't require newbies to buy anything to get started. They ease them into buying currency by first giving them plenty of free things to do. As they do specific tasks, the player earns in-game currency. The player then discovers that the virtual cash unlocks more enjoyable experiences in the game. Soon, they find themselves desiring virtual money and the virtual assets that the money can purchase. That creates virtual demand.
Those early tasks that onboard you into the virtual economy include things like cutting treats in a virtual forest or slaying low-level monsters. For example, in the popular MMORPG World of Warcraft, killing low-level monsters is a low-risk way to earn "gold," the name of the World of Warcraft currency.
But soon, that player will find those virtual cash-earning tasks to be rote and time-consuming. Soon it feels easier to spend $10 buying some virtual cash than spending a real-world hour cutting virtual trees before long.
That's how we got the phenomenon of gold farming, where players pay gig workers to do these tasks for them. The gamer hands their account over to a gig worker who then does the virtual tree cutting and mineral mining while the gamer is sleeping or working or otherwise fulfilling their pesky real-world responsibilities. It is better to pay a gig work $1 for X units of virtual money than pay the game company $10.
Neal Stephenson foresaw crypto gold farming.
I mentioned two Neal Stephenson books with heavy crypto and virtual world themes: The Diamond Age and Cryptonomicon. There is a third, lesser-known book called REAMDE.
One of REAMDE's themes is designing virtual world economics. The protagonist is a technolibertarian entrepreneur piqued by how game companies try to block gold farming because it threatens their in-game currency sales. He designed a game that embraced gold farming and made it easy for players to turn virtual currency back into real-world cash. He went so far as to hire a geologist to design realistic spatial distribution of natural resources in the virtual world so that the world's geopolitics would evolve to resemble that of reality. In the plot, this naturally leads to the game becoming immensely popular.
Indeed, many people make a living in gold farming. Full-time gold farmers often reside in developing nations. There are outsourcing companies that employ armies of gold farmers for fixed wages in places like China (at one point, there was reporting on gold farming with prison labor).
These people lose their ability to make a living when the game companies or governments crack-down. That creates an opportunity for games with a blockchain-based virtual economy. For example, gold farming and cryptocurrency has arisen in Venezuela, where gold farmers take payments Bitcoin instead of devalued local currency.
Axie Infinity is one instance I've found for Stephenson's vision in REAMDE.
Axie Infinity is an NFT game made by a Vietnamese startup. The game resembles Pokemon in that players collect and battle cute creatures (Axies). Battling Axies yields a crypto token that is used to create new Axies. Gold farmers in the Philippines gold farm this token to sell on crypto exchanges to people who wish to make more Axies.
Some in the crypto tech community, inspired by Axie Infinity’s empowerment of Phillipino gold farmers, are designing more “play-to-earn” games, hoping this will address the onboarding problem I highlighted in my previous post. What better way to attract new users than by paying them to kill monsters?
So what happens when people use AI to gold farm in crypto games?