> home
> About
> Contact Us
> Editorial Info
> IEEE-USA

04.10

Backscatter: Don’t Blame Us

BY DONALD CHRISTIANSEN

Here is the news I’ve been waiting for. It turns out that when technology goes bad, it may not be the fault of engineers after all. Or even of technology’s users.

It seems that the time-honored notion held by many members of the general public, namely that faulty technology or its misuse is primarily the responsibility of its inventors and developers and secondarily of its users, is now being questioned. It could be inanimate technology itself that may be the culprit, or at least a co-conspirator.

Without mentioning the controversial topic of singularity, certain thinkers are strongly suggesting that machines must already shoulder responsibility for unfortunate outcomes in which they are involved.

Supporters of this theory attempt to deflate several conventional arguments against it. The traditional argument that must be overcome is that a machine is not human, and so cannot be held morally responsible for its actions. A relatively weak response intended to defuse that challenge is that certain non-human actors (e.g., corporations and some animals) may also be held accountable for their actions. Yet both have an element of consciousness that machines do not.

A second argument that must be refuted is that a machine is created and operated by humans, who thus are responsible for all its actions. One counterargument put forth is that a computer system, for example, can do things that were not contemplated by its designers or users. “Things will have their own way,” as one writer put it, sympathetically calling the machine’s victims “hapless humans.”

Lest you may be wondering whether I am swayed by any of these counterarguments, let me make it clear that I am not. Every machine or complex system I’ve ever met turned out to be dumb. If any appear to be inherently intelligent, they are faking it—responding for better or worse with actions whispered in their ears by humans.

Having said this, I nevertheless humor my own computer. And when it misbehaves, I chastise it, sometimes severely, making sure no young children are within earshot. I rat out my computer, revealing to my colleagues when it is behaving badly, running slowly, or refusing to carry out my wishes.

Machines on Trial

Another proposition by those wishing to bring machines before a grand jury, indicting them for their misdeeds, is given again in the example of a computer program. A computer is constrained in what it can and cannot do, and thus it limits or otherwise mediates what users would like it to do. If used to help make medical diagnoses, it will only accept inputs that are predefined and in a specific format. The machine-accusers assign this limitation to the computer itself, not to its programmers. Users in the medical profession are seen, perhaps, as codefendants for accepting the limitations imposed by the machine and its diagnostic program. More consequential, perhaps, is the notably overoptimistic belief that when some technology is incriminated in some immoral action, as in the case of napalm or cluster bombs, it could be sentenced to “execution.”

If there is a redeeming aspect to the philosophy of those who believe in machine consciousness and responsibility, it may be in their encouragement of human brain research. It may also to some extent reinforce failure analysis techniques. As one believer in machine-guilt put it, “The point is not to absolve humans of responsibility, but to attribute responsibility where it lies, regardless of humanness.”

Of course there are the singularity extremists who stretch the notion of machine consciousness, yearning for the day when they themselves can roam the earth as a reconstructed clone. Its artificial brain will be loaded with their life experiences. If you count yourself among those dreamers, let me remind you that your clone would not be you. It would be a total fake. The unfeeling robot would not experience pleasure or pain, nor be conscious of the feelings of any living creature in its social circle.

I ask that you not mention any of the foregoing to my computer. I do not want to hurt its feelings any more than I have already done through my incessant complaining about its shortcomings and intentional errors.

References

Arnold, M., and C. Pearce, “Is Technology Innocent? Holding Technologies to Moral Account,” IEEE Technology and Society Magazine, Summer 2008.

Perkowitz, S., Digital People, Joseph Henry Press, 2004.

Birsch, D., “Moral Responsibility for Harm Caused by Computer System Failures,” Ethics and Information Technology, Vol. 6, 2005.

Floridi, L, and Sanders, J.W., “On the Morality of Artificial Agents,” Minds Much, Vol. 14, August 2004.

Horgan, J., “The Consciousness Conundrum,” IEEE Spectrum, June 2008.

Nissenbaum, H., “Computing and Accountability,” Commun. ACM, Vol. 37, No. 1, 1994.

Swierstra, T., and J. Jelsma, “Responsibility without Moralism in Technoscientific Design Practice,” Science, Technology, and Human Values, Vol. 31, No. 3, 2006.

Donald Christiansen is the former editor and publisher of IEEE Spectrum and an independent publishing consultant. He is a Fellow of the IEEE. He can be reached at donchristiansen@ieee.org.