Frankenstein: the Modern Prometheus (Book)
By ai-depot | June 30, 2002
Creation of Life
The primary purpose of this book is to analyse the human desire for creating life, and the consequences they bring upon themselves by playing god in such a way. To this end, Shelley tells the exhilarating tale of a talented and ambitious scientist, who blindly decides to create a human-like life-form. This last point seems to be one of the few areas not covered by the book: Frankenstein does not even consider the consequence of this actions beforehand. So to that extent, Shelley contributes to the discussion by extrapolating on potential consequences of the creation of life.
As mentioned in the introduction, this story fits in extremely well in modern times, with some biologists threatening to clone humans. From the point of view of Artificial Intelligence, this discussion seems a long way off. However, discussing the possibility it seems a wise idea for the sake of prevention, if nothing else.
Ethical Standards
Very recently, as you may know, Britain has been forced to pass an ‘emergency’ law preventing an Italian doctor to clone a human within the UK. Many arguments have been put forth to support this, involving culturo-social consequences (such as discrimination), pollution of the gene pool, human rights and legal issues, and potential personal problems of the clones. Many opposing arguments can be conjured up in a similar fashion, which clearly identifies the need for standardised ethics.
For the creation of virtual lives, the matter is different. The creatures would interact with humans in a completely different fashion, which seems less important due to its virtual nature: at the press of a switch, “Nothing ever happened.” Nevertheless, the impact of an intelligent virtual creature with personality issues may be colossal, in this world increasingly dependant on inter-telecommunication and digital computing.
So, similarly to its biological counterpart, the creation of virtual lives seems to also require a set of ethical standards. However, this would be much more challenging to establish and enforce, since it deals with such cutting edge technology. Any commission in charge of setting the standards would fall behind current work, thereby desperately trying to play cat and mouse with researchers.
Responsibility of the Creator
This implies that researchers creating virtual lives need to be responsible for their creatures. This is crucial not only for preventing potential problems, but also from a legal point of view. Creating a law which pins the blame of a virtual being’s actions on the creator would force him to think about the consequences before ‘birth’.
Once again, this is much more difficult to establish, and enforce. Firstly, if the virtual being is that smart, how can it be traced to its roots unless it cooperates? A fair amount of activity logging would partially remedy this problem (server activity logs, CCTV cameras). Secondly, how is the being to be neutralised in the first place? There’s only so much damage a mammal can take, but a rogue agent on the internet taking advantage of bugs and loopholes would not be controlled out that easily.
Even once such measures are taken, what prevents ‘rogue’ A.I. researchers from not adhering, thereby voluntarily seeking fame and fortune wherever they are given a free hand? Sadly, very little stands in their way, just like for the cloning debate.
Research Context
I thoroughly believe that the first intelligent creature will be created by accident. Either by a teenager who has hacked together a few rules capable of being expanded, or by a long winded evolution on a super-computer. However, for those worried about the potential of their research and who set out to create a virtual life in the first place, there is a need for a safe development environment. By this I mean a bullet-proof sandbox, which any artificial being would not be able to cause harm. This is crucial for preventing gigantic damage to the internet for example, but also to allow us to press the ‘Remote Detonate’ switch if things get out of hand. This is always a necessity, as no amount of human planning can anticipate all scenarios, as much of history has revealed.
Tags: none
Category: review |
