London, 54p7
So I moved to London to attend the Royal College of Art’s Design Interactions program with Anthony Dunne, Fiona Raby, Brendan Walker, James Auger, Nina Pope and a host of very different, motivated and bright students. The feel of the school is pretty different to UDK, probably because (even though it is in central London) it operates more of a campus which includes most facilities you would need throughout the day, including a fun bar. This way, people are being exposed to each other all the time, bonding them together much more strongly than I am used to and facilitating constant exchange between them. I actually prefer it this way, I believe.
Bruce Sterling came over for a talk on Monday and this also launched our first project, the brief of which is titled 54p7. It is about re-imagining robots and it is very open to interpretation. What really struck me in Bruce’s talk was his emphasis on how the idea of the holistic robot has proven to be a failure, with machines still struggling to understand basic language, let alone understand their surroundings. What arises from this is the image as the robot/computer as a somewhat disabled entity, maybe rather autistic because they tend to have what is often called splinter skills, such as extreme precision, speed or obviously computation.
To me this makes clear that machines, at least in the nearer future, will be highly dependent on human reasoning for most things outside their area of expertise. Humans (if you exclude actually autistic individuals) however, tend to be the opposite. Very good at subjective decisions (”I like”), but rather bad at moving around car bodies with precision or crunching PI. Consequentially, an interesting thing to propose might be a symbiosis between the human and the computer which, if you want, would result in a kind of much more powerful robot. Of course you could argue that this already exists in the form of different machines which extend one’s power, etc. but I am actually more thinking of an exchange between two separate entities.
I’ve been thinking hard about how this exchange might exactly work, and also tried to re-evaluate the general relationship man/machine for a second. The classic 20th century idea of a roboted life would mostly include a paradise with the machines doing all the work and humans having a good time in the park. From today’s perspective and our experiences with increasingly globalized work-processes, this prospect seems less promising as machines put people with replaceable jobs out of work due to their greater economical efficiency. Today’s ideal (in the western world) is the all-brain society which just thinks up great concepts and sends them to either the machines or South-East Asia for completion. Yet, there will hardly ever be enough clients or such jobs in general to cater to the 700+ million of, for instance, Europe. At some point, we might actually need machines to generate work for people in order to keep them alive or at least busy. So what I am thinking of right now, would be the perversion of the original idea of the robot towards a machine which rather produces work, especially tasks that it cannot perform itself, such as value assessments, moral judgements, aesthetic decisions, etc. I am currently figuring out the mechanics of that relationship, but a scenario could look like this:
An individual gets fired by his or her company because his or her (probably manual) job has been cut or moved to somewhere cheaper. He or she goes to the local employment center and after assessing skills and finding out that there’s nothing currently available in that sector, gets offered to participate in a companion program. This program consists of a little computer which easily could be fit onto (into?) one’s body and is so energy efficient that it can actually be powered by body heat and/or movement. The person wears that machine and every time it would get stuck and needs human input, he or she logs on to a service very much like Amazon Mechanical Turk and completes a few tasks as outlined above for which he or she will be payed in return. As said, I’m still trying to figure out the exact nature of this symbiotic relationship, but the prospect of commenting on social issues by inverting the paradigm of the robot itself seems kinda promising.
October 15th, 2007 at 6:35
What interests me about this suggestion is the way in which it would subvert the illusion of neutral tool-like utility is already at work in the exchange between human and computer, and which fuels the robotic servitude fantasies alluded to above–which reminds me of a song by R.Crumb’s band the Cheap Suit Serenaders called “On My Persian Rug” that has the line “robots bring me gifts of wealth and pleasure, slave girls sing to me love songs ever”. In turn, language itself prepares us to participate in this: tricking us into thinking that it is a relatively neutral medium for the exchange of information that originates somewhere before language, before discursive thought. My fuzzy little point here is that by inverting the expected nature of the exchange between man and machine, you interrupt the ability of that exchange to convince you that it moves in only one direction, that there is only one user and one used, and that AI/robot/machine is principally a means to intervene in the world, to manifest some intention or desire that originates somewhere before the technological exchange–rather than being somehow, perversely, a product of it.