The rise of the techno-optimists

By Prasan Samtani

See, I have this day set thee over the nations and over the kingdoms, 

to root out, and to pull down, 

and to destroy, and to throw down, 

to build, and to plant.

Jeremiah 1:10

If there is a polar opposite to the spirit of Jeremiah, it is to be found in the spirit of “The Gundo” (referring to the city of El Segundo, the spiritual Mecca of the techno-optimists). Unlike the doomsayers who presage the end of modernity and planetary overshoot, the hard tech entrepreneurs of The Gundo have an infectious optimism: their worldview is straightforward - we are in despair because we forgotten how “to build”, and the solution is not to discard or adapt modernity, but to accelerate it.

On the surface, the philosophy couldn’t be more opposed to doomer-optimism:  maximize energy production (Valar Atomics), eliminate humans from manufacturing (Freeform, Machina Labs), large scale carbon removal (Terraform Industries), weather control (Rainmaker), and of course, the quest to become an interplanetary species (Relativity, K2, and the grandaddy of them all, SpaceX). 

A Falcon 9 launch.

Wendell Berry’s essay, Faustian economics, argues explicitly against this “credo of limitlessness”, describing it as a kind of “autistic industrialism”, where knowledge is constantly advanced without guiding principle.

Knowledge is as food, and needs no less

Her temperance over appetite, to know

In measure what the mind may well contain;

Oppresses else with surfeit, and soon turns

Wisdom to folly, as nourishment to wind.

John Milton, Paradise Lost

The spirit of The Gundo however, demands that limits be explicitly disrespected. If the Earth is insufficient for mankind, so be it. We shall go beyond.

Yet, there is much to admire - first, unlike the weasly “Davos Man” who speaks of economic decoupling while globetrotting on his private jet, the “Gundo Bro” does not engage in this kind of self-deception. Their diagnosis is not that we must reject limits while they are free to transgress them - but that the limits must be shattered through the conscious application of human will.

There is also more than a kernel of truth in the belief that we must redevelop the ability to produce the material basis of our civilization domestically. But here they miss a crucial thing - it is not enough to “make America build again”. It is just as, if not more important, to “make Americans build again”. Manufacture and production have both technical and social components - redeveloping the ability to do the former without attempting to (or in fact subverting the ability to) do the latter cannot produce the national spiritual revival required to shift away from a consumer society. A man makes an idol from hammer, chisel, and rock; the hammer and chisel in turn make man into sculptor. A nation with grand statues but devoid of sculptors is missing something, and that is not remedied by technical progress alone.

Then, there is the ultimate question - what purpose does the technical advancement serve? Even if we are to transgress natural limits (which I believe we must), there are undoubtedly reasons that provide good motivation for doing so, and others that seem utterly wasteful by comparison. For example, very few people can convincingly argue against the development and transfer of agricultural technology during the Green Revolution Period (1940-1980), which saved hundreds of millions of lives globally. Or against genomic advancements which are now entirely reliant on high-performance computing, but have the prospect to radically improve the condition of people who are otherwise doomed to suffer lives of misery and disease.

Other reasons seem trivial and wasteful from the outset - for example, Meta recently announced plans to purchase 350,000 H100 GPUs to add to their GPU fleet for reasons that seem mostly to do with entertainment. For context, an H100 has a max power consumption of 700W, which means the new additions to their fleet will consume about 245MW of electricity (and there are sound economic reasons for them to try and maximize utilization, indicating that this is a reasonable estimate of the power draw). The added power utilization alone matches the approximate power output of the natural gas power plant that my house neighbors.

To power 350,000 of the machines on the left, you need one of the plants on the right.

Far from decoupling, it seems fairly likely that we could enter an era where large scale computing companies will find it worthwhile to build out utility scale power generation facilities to adequately supply themselves with energy for AI related tasks.

In his fantastic book, “The Art of doing Science and Engineering”, Turing award winner Richard Hamming gives us a visual of a drunken sailor at a party. Undirected, he stumbles back and forth, getting nowhere. But at the sight of a pretty girl across the room, he ambles his way over to her, even if a few steps along the way are less efficient than if he were sober. Without a guiding vision, an overriding push for “more” leads us away from asking questions about how we would like to organize society more broadly - both physically and socially, for the two are tied together with a single yoke.

Will the new tools be as the chisel, the plane, and the lathe - which allow man the rest of contentment after seeing the good work of his own hands? Or will they, in the words of Russel Kirk, be as mechanical Jacobins, alienating man from place and society?

The trouble is, they have no needs. You get everything you want just by imagining it. That’s why it never costs any trouble to move to another street or build another house. If they needed real shops, chaps would have to stay near where the real shops were. If they needed real houses, they’d have to stay near where builders were. It’s scarcity that enables a society to exist.

C.S. Lewis, The Great Divorce

About the author:

Prasan Samtani (dhanuraashi@) is a Software Engineer at Google living in Pasadena CA; father, husband, and a hobbyist gardener, baker, & carpenter.