What is being promised by the automations associated with AI and other emerging technologies … and what are we expected to give up in return?

**

In their article “What tech futurists get wrong about human autonomy”, Divya Siddarth & Kelsie Nabben (2021) make a compelling argument that the push toward technological automation is underpinned by misplaced (if not disingenuous) promises of autonomy.

On one hand, is the idea of the autonomous machine – promising an ‘autonomy of outcome’ where technology is capable of reaching effective and efficient outcomes with ‘minimal active decision-making input’ on the part of humans. This implies technology that has “become self-governing, self-defining, self-learning and self-directed”.

Many people will see little reason to contest this form of technological autonomy, even if it does mean that the ‘traditional’ forms of human governance and oversight are sidelined, if not usurped, altogether. As Siddarth & Nabben put it:

“The clear superiority of machine decision-making in this imagined future makes human self-governance laughable. Of course we would choose the perfect, quasi-omniscient decision maker over our own imperfect selves”.

Alongside this promise of autonomy of outcome, then, is the corresponding promise of humans benefiting from increased “autonomy from need” — i.e. “a freedom from the daily toil of being human”. In this sense, technology offers humans a welcome mental relief, what Siddarth & Nabben describe as the opportunity to “trade human autonomy for an autonomous future free from the burden of labour”.

Again, this form of technological autonomy is also likely to be a welcome development for many people. Under these conditions, we no longer need to endure the toil of having to take routine, boring decisions. We no longer need to worry whether we have done the ‘right’ thing or not – whether we have missed something important, or similarly ‘slipped-up’. We no longer need to feel accountable for our fallible memories or moral judgements. Instead, we can be ‘freed up’ for other (less decisive) activities.

Of course, things are not as simple as this. As Siddarth & Nabben put it, these visions of technology-induced autonomy are founded problematically on the “mistaken belief that automation can achieve any meaningful autonomy at all”. Instead, as Siddarth & Nabben point out, genuine human autonomy is not an individual trait or quality, but something that is circumscribed collectively by the communities and societies in which we live. One cannot simply decide to be ‘autonomous’ – instead, individual and community self-sovereignty is borne from social relations with others.

Seen from this perspective, then, the visions of technology-led autonomy just outlined seem likely to simply diminish, degrade and sideline the aspects of humanity that are capable of giving humans genuine autonomy. These include conditions of cross-community dependence, space for disagreement and deliberation, and human higher-order critical thinking in cooperation and coordination. 

Perhaps most pernicious is the promotion of AI and other automated technologies as somehow capable of transcending the political – offering humans a life free from having to address collective problems through an explicit politics, and, instead translating matters of political judgment into questions of technical expertise. Again, Siddarth & Nabben contend that this is a false trade-off, and that politics is an essential element of what it is to be human. 

In short, these are not ‘conveniences’ that any flourishing human society should be looking to encourage. Indeed, these technologies depend on massive centralisations of power in order to guarantee their ‘success’ – not least the concentration of massive amounts of data, computers and capital that we are already seeing in the rise of Mega, Alibaba, Alphabet, Amazon and other tech giants.

All told, there is very little in the promises currently sustaining popular enthusiasms for automated technology that seems to be particularly advantageous. Any resulting human autonomy should be considered rather hollow reward in light of all that we are expected to give up in return. As Siddarth & Nabben contend, the fundamental flaw in all this talk of human and machine autonomy is that it “takes for granted that the goal of autonomy thus presented is worth pursuing”. These are promises that need to be resisted rather than embraced.

**

REFERENCE

Siddarth, D. & Nabben, K. (2021). What tech futurists get wrong about human autonomy. Noema , 9th December, https://www.noemamag.com/ai-blockchain-human-autonomy-future/