Progress is a Byproduct

There is a faint sense of guilt that follows the act of wanting. Wastefulness, luxury, living beyond one’s means. Frugality is treated as a virtue.

Let’s talk about Ajinomoto and gaming.

I. From Kelp Stock to Semiconductor Components

In 1908, a chemist named Kikunae Ikeda extracted a single substance from kelp stock: glutamic acid. He realized there was a taste that was neither sweet, sour, salty, nor bitter, and he named it “umami.” The following year, it was marketed as Ajinomoto.

The motivation was simple: the desire to eat delicious food, cheaply and in abundance. While it is often described as the discovery of the fifth basic taste, that isn’t why it sold. It sold because it made cheap ingredients taste better.

A hundred years passed from that point.

The semiconductors inside today’s PCs and servers have a structure where thin wiring is stacked in multiple layers. Between these layers, an insulating film must be inserted. One of the materials used for this film is called Ajinomoto Build-up Film. It is manufactured by Ajinomoto. The organic chemistry developed to produce amino acids in massive quantities at low cost translated directly into resin technology, which then became a component for computers.

II. The Desire to Play

A company founded in 1993 called NVIDIA released a component called the GeForce 256 in 1999. It was a device for rendering 3D images on a screen, and it was the first in the world to be marketed as a GPU. It excelled at performing massive amounts of the same calculation in parallel, and its purpose was singular: gaming.

It sold. Because it sold for the sake of games, volume increased, prices dropped, and performance rose. With every generation, it became faster.

In 2006, NVIDIA released a framework called CUDA. This allowed GPUs to be used for calculations other than drawing images. Researchers began using it.

In 2012, at a competition for image recognition accuracy, a system trained using two GeForce cards won by a landslide. From there, deep learning accelerated rapidly to where it is today. Almost all the artificial intelligence currently operating exists upon this lineage.

III. No One Aimed for This

Both cases happened in the same order.

First, someone wanted something. Money was paid to satisfy that desire. Because money accumulated, technology in that direction advanced. That advanced technology was then used in a completely different place.

It wasn’t that the technology was conceptualized first and then funded. The desire came first, money gathered around it, and the technology grew in its wake. It wasn’t a national plan or a researcher’s vision.

And no one chose the destination. The person who wanted to extract stock didn’t pay money for semiconductor insulating films. The person who wanted to play games didn’t pay for artificial intelligence. To the people paying, it was simply about delicious food and play.

What becomes possible is only understood in hindsight.

Wanting Properly

There is perhaps only one conclusion to be drawn from this.

Wanting things properly leads to the ability to do new things. At least, that is how it looks when looking at history.