Ours is the age teeming with intelligence. We have or will soon have the ability to manufacture anything in the world of bits. A straightforward projection of trends augurs well for production in the world of atoms, too.
But real abundance — the sort that matters to our minds, our bodies, our relationships, our environment — does not follow inexorably from sufficiently-voluminous intelligence. Securing it requires continuous application of aligned judgement: the use of our human faculties to evaluate each product of intelligence (a new policy, widget, agreement) against that future which we, as a collective, have set our sights on. It is not solely the magnitude of the intelligence vector, as it were, but its direction.
So what, exactly, is that future? How do we select among a multiplicity of options? And once we’ve determined what future we prefer, how might we actualize it?
The future worth building is human. It will be constructed through human-made processes, it will be made with human-like qualities, and it will be formed using human-controlled tools.
In so articulating, I reject a machine future. I repudiate the notion that humanity’s role is merely to bring a new, designed, successor species into the world. The transhumanist future is an inhuman one. Humans are not means to an end; we are ends in ourselves.
The headwinds are substantial. The state is keen to hoard advanced artificial intelligence for its security. Markets will naturally want to commoditize it, and market participants will want to monopolize it. The individual human, acting alone or in concert with other engaged humans, will struggle against these incumbent hegemons. In other words, our present course may very well centralize control of AI in ways that flatten human difference and cap human potential.
Our way to a maximally-human and human-maximizing future is through the human work of politics. Human actors acting freely — not government actors, and certainly not corporate actors — are uniquely capable of this. Humans as individuals are permitted to hold and develop normative perspectives that can shape the future, and they elude formalization. Humans’ emotional capacities are as much of an asset as their rational ones. The subjective is on equal footing with the objective.
I look forward to helping build this human future. As a congressional staffer, I have a small role to play. Legislation is important: the Congress will need to put together an unprecedented set of good policies that spans catastrophic risk, AI liability, workforce, and taxation, among others. But one landmark statute won’t make the future, far from it. Nor will one blueprint, one convening, or one alignment technique. The road ahead will instead, to quote John von Neumann, involve “a long sequence of small, correct decisions.” We should start making some today.