On his Amor Mundi blog, which I read regularly, Dale Carrico, a lecturer with the Department of Rhetoric at UC Berkeley, frequently rails against what he calls Superlative Technology discourse.
For instance, in a recent post, which begins by saying:
“I always cringe whenever some “inspirational” business huckster barks into the mic or at the dinner table about how “innovation,” “production,” “enterprise” or whatever slogan the Amway mindset has latched onto at the moment has “no limits, man!” so, you know, we should all presumably become dot-eyed marauding maniacs and “go for it!” and “take no prisoners!” or what have you.”
He goes on to criticize:
“Superlative technocentric fantasies of selves reduced to streams of spiritualized digital data (a formulation preceded by and prepared for by decades of reductive accounts of selves reduced to expressions of genetic information)”
“reduced “instantiations” of software instructions by angelic nanoabundance (a formulation preceded by and prepared for by, among other things, decades of PR accounts of commodities as indifferent sites for the distinguishing emblemization of designer logos)”
“deliberative politics reduced to the consequentialist computation of “optimal” outcomes by Robot Gods (a formulation preceded by and prepared for by the anti-democratic Cold-War politics of neoliberal incumbency, with its corporate think-tanks, technocratic military experts, and corporate broadcast-mediated manufacturing of consent)”
To say right away, I completely agree with him that the idea of “no limits” is immature and oversimplistic. There must always be limits, otherwise there is chaos. Not top-down instituted limits, but bottom-up and organically implemented limits.
To simplify, the three technologies he criticizes are:
- mind uploading
- molecular manufacturing
- superintelligence
To respond to each briefly: if functionalism holds, and computer speed, memory, and brain scanning resolution continue to improve, then the feasibility of mind uploading is highly likely. If functionalism does not hold, then mind uploading isn’t ruled out, but you might have to get more creative to implement it. (I.e., running a 10^11 neuron biological brain on a 10^20 neuron biological substrate with greater flexibility.) Functionalism is defined by Wikipedia as follows:
“Its core idea is that mental states (beliefs, desires, being in pain, etc.) are constituted solely by their functional role — that is, their causal relations to other mental states, sensory inputs, and behavioral outputs. Since mental states are identified by a functional role, they are said to be multiply realizable; in other words, they are able to be manifested in various systems, even perhaps computers, so long as the system performs the appropriate functions. While functionalism has its advantages, there have been several arguments against it, claiming that it is an insufficient account of the mind.”
Therefore, arguments against the feasibility of uploading should be directed against the philosophy of functionalism, not the supposed motivations of those who believe uploading is feasible, such as myself.
In other words, attack ideas, not people.
To address the concept of “angelic nanoabundance”, this is what I would call superabundance — the notion that, given reprogrammable factories that can self-replicate from simple raw materials and (preferably solar) energy input, great material abundance could be achieved quite quickly. Nanotechnology is not even necessarily required to build such a device. RepRap, a project whose motto is “Wealth without money”, is a nascent example of such exponential manufacturing capability.
If exponential factories are indeed possible, then practical superabundance is quite likely. A single 1-kg factory, self-replicating once per day, could produce approximately two billion kilograms worth of factory in just a month, which could then be used to produce housing, transportation devices, infrastructure, electronics, and so on, in similar quantities in just one production cycle. This technology, if distributed widely and regulated intelligently, has the potential to wipe out poverty once and for all. The humanitarian potential is enormous, and this was acknowledged by several speakers at the recent CRN conference, including Linda Hopper, the founder of WorldCare, a leading charity.
No designer logos involved. Molecular manufacturing, as well as less sophisticated manufacturing technologies, could facilitate great abundance, and also great danger. Think about mass producing missiles that cost practically nothing.
To address the third item, superintelligence — whenever anyone talks as if Friendly AI or any other form of superintelligence will facilitate an “optimal” outcome, they’re being ignorant. (This includes anyone with the Singularity Institute who has ever said such a thing, though I don’t believe I’ve ever heard it implied.) Life is too complex and multi-faceted for there to be a thoroughly optimal outcome for anything. Preferences constantly shift, and satisfying a complex web of interacting preferences in a dynamic world is a huge task for any intelligence.
There could indeed be robots, or cybernetic organisms, or artificial intelligences, created in the next century that have powers some humans might consider “god-like”. This is a natural consequence of the fact that biological neurons only operate at 200 Hz and synthetic neurons could operate much faster, and that robotic arms can theoretically operate much faster and with more agility than the human motor system. Coupled with exponential manufacturing, a cybernetic organisms capable of distributing its intelligence could make itself extremely capable of influencing the external world, even to the point of being able to wipe out the human race — or greatly improve our lot. People are free to take this scenario seriously or dismiss it as they like, but there is an ever-expanding dialogue among thinkers concerned that AI or augmented beings could indeed exceed human intelligence and capability in the 21st century. Once they exceed it, they will necessarily soar past it, fueled by a positive feedback loop of self-improvement.
No relation to Cold War politics. If we want to see a democratic future, then we have to construct superintelligence in such a way that it cares about what other people think. For instance, if you’re working on a Brain-Computer Interface project to create a superintelligent human being, use an altruistic human being instead of an egoist one.
To quote a further passage from Carrico… I’m trying to post it as concisely as possible, but the following is literally just two sentences:
“If the Superlative Technocentrics were actually right to imagine that billions of people now living will find themselves all too soon living in a future transformed by Friendly or Unfriendly post-biological intelligences, nanotechnological superabundance, biomedical immortality, or the like (and I do think they are far more likely to be wrong than right and I think this matters enormously), even granting them this, I think they are profoundly wrong to imagine that our best way to facilitate the best, least violent, most fair (or whatever) versions of these Superlative outcomes is to contemplate and prepare for the Superlative outcomes themselves, in the abstract, as these outcomes suggest themselves to us in our own impoverished vantage (an impoverishment exacerbated all the more by marginal and anti-democratic modes of Superlative deliberation). Such contemplation and preparation circumvents the ongoing and plural stakeholder contestation that will certainly articulate the unpredictable developmental forces and the dynamic developmental pathways along which such outcomes would actually “arrive” (were they to do so), ignores the practical, scientific, technical, pedagogical, regulatory, cultural knowledges arising out of our collective day to day responsiveness, competition, and collaboration in the plural presents from which no less plural futures will present themselves, that will not only shape but actively constitute our foresight and provide the living archive to which future generations or the communities in which we will ourselves later belong will make our collective recourse as we struggle to cope with these outcomes and their alternatives.”
My only response here is to say that I invite all stakeholders to contribute the technological, social, and regulatory-proposal dialogues ongoing in the relevant communities.
I strongly identify as what Dale would call a Superlative Technologist, in that I believe the advanced technology of the 21st century will facilitate sweeping, “superlative” outcomes, whether we want it or not. The only way I see to channel the coming flood is to acknowledge its likelihood and do our best to ensure that the initial conditions benefit all of humanity rather than some self-identified “elite” subset.
41 Responses »
