Or, rather what it really means to be human?
We create in our own image. Or do we? The projects researched here were used by Lucy A. Suchman as examples that: ” make evident how roboticist imagine humanness”.
In the case of the human, the prevailing figuration in Euro-American imaginaries is one of autonomous, rational agency, and projects of artificial intelligence reiterate that culturally specific imaginary.
The text derives upon the chapter 13: Figuring the Human in AI and Robotics, from the book by Lucy A. Suchman: Human-Machine Reconfigurations, Cambridge University Press, 2016.
Creating in our own (rational) picture?
I adopt the view here that we need to include in our analysis the question of just what constitutes agency in any case, for humans or nonhumans. Efforts to establish criteria of humanness (for example, tool use, language ability, symbolic representation) have always been contentious, challenged principally in terms of the capacities of other animals, particularly the nonhuman primates, to engage in various cognate behaviors.
The 3 ingredients of humanness
Lucy considers 3 elements that are “necessary for humanness” in contemporary AI projects: embodiment, emotion, and sociality.
And with thorough analysis comes to the conclusion that the most important feature of an AI system is not its ability to discover and react to ‘classified’ emotions or its ability to have a ‘body’ and react to the space.
But First – Stelarc’s Prosthetic Head
This is not an illustration of a disembodied intelligence. Rather, notions of awareness, identity, agency and embodiment become problematic. Just as a physical body has been exposed as inadequate, empty and involuntary, so simultaneously the ECA becomes seductive with its uncanny simulation of real-time recognition and response. Initially decisions would have to be made about its database and whether The PROSTHETIC HEAD is a pathological, philosophical or simply a flirting head. A problem would arise though when the PROSTHETIC HEAD increases its database, becoming more autonomous in its responses. The artist would then no longer be able to take full responsibility for what his head says. http://stelarc.org/
The author visited Stelarc’s exhibition in Canada in 2003 and had a valuable encounter not only with the artwork but with the author himself. What she has found to be interesting for her story is that the design of the Head was, in a sense a very human “idol” (literally in terms of pagan rituals) of a human, and its capability to anticipate the relationship in a conversation and to interact as well as leave the questions “open” and almost make a joke.
Andrew Pickering, cited by Lucy, in his book “Cybernetics and the Mangle” (2002) gives a more precise view of agency – “the liveliness” of the non-human agent or its ability to “reconfigure itself in response to its inputs”. In comparison with a lab-box-bound Kismet and its close relative Cog, if we put aside the fact that their softwares got rusty because the research stopped, the head was equipped with social skills that were far more reaching then the book of proper emotions on which the other two robots were trained at.
There is a new light in how we can observe creating new AI systems. I do not create with “me” in mind. We create with ‘us in mind – “us” socially, historically, spatially, habitually. The ‘intention’ as one of the constituents of agency is also redefined since the mind is not observed as only within one person but in terms of the relationship and cumulative historical and social map the ‘non-human’ and human belong to.
What if we understand persons as entities achieved only through the ongoing enactment of separateness. Rather then working to create autonomous objects that mimic Cartesian subjects, we might then undertake different kinds of design projects…
And while researching Stelarc’s head photos, I have come across this intriguing piece of interconnected humans and non-humans by Erin Gee: Orpheus Larynx – Vocal performance feat. Stelarc and his “Thinking Head”. Performed live at Powerhouse Museum of Technology, Sydney AU 2011.
It was a choir of 3 heads and further on, some robots and one human with a device. Erin writes on her website:
I sang along with this avatar-choir, while carrying my own silent avatar with me on a digital screen.
It is said that after Orpheus’ head was ripped from his body, he continued singing as his head floated down a river. He was rescued by two nymphs, who lifted his head to the heavens, to become a star. In this performance, all the characters (Stelarc’s, my voice, Orpheus, Eurydice, the nymphs) are blended into intersubjective robotic shells that speak and sing on our behalf. The flexibility of the avatar facilitates a plurality of voices to emerge from relatively few physical bodies, blending past subjects into present but also possible future subjects. Orpheus is tripled to become a multi-headed Orpheus, simultaneously disembodied head, humanoid nymph, deceased Eurydice. The meaning of the work is in the dissonant proximity between the past and present characters, as well as my own identity inhabiting the bodies and voices of Stelarc’s prosthetic self.
*the header image – Jordan Wolfson’s Female Figure (2014) from MildHangover on imgur
*more on Jordan Wolfson https://www.davidzwirner.com/artists/jordan-wolfson