3 Laws and Drones

Science fiction has pointed out fears concerning AI and various apocalyptic scenarios; this is the job of the creatives. Perhaps the most famous has been Isaac Asimov's books on robotics in which he established the following 3 laws to protect humanity (from here):-

A robot may not injure a human being or, through inaction, allow a human being to come to harm.
A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

In humanity's development of AI so far, have we followed these laws or similar? And the answer unequivocably is NO.

And the problem is racism, humanity does not have respect for the life of the other. And where does the problem lie? With the power in the West. It is the nationalism of the West that is the problem, a nationalism that is prepared to use any weapons technology possible to further its own interests.

The main despicable weapons technology that uses AI is drones, and NATO's use of drones is shameful. These are unmanned crafts that are being used to murder people in territories where there are no westerners. If you look at Asimov's laws then western use of drones breaks the first 2 laws.

This is the problem with AI, if it is not programmed for safety; here is a Reuter's article which says just that. The article is hopeful because it suggests that appropriate protective programming will be in place. But they are not facing the fact that AI is being used as weapons already as in drones and other technology. Protection is not in place already.

In other words these people have their heads in the sand:-



You could argue that drones are fired by soldiers in the Nevada desert but according to Asimov's 3 laws that would not be allowed. The drones are killing, the drones are AI. Smart bombs are killing, smart bombs are AI. Applying Asimov's 3 laws neither drones nor smart bombs would be able to kill even though humans are pushing the switch.

Protective laws need to be universal. They need to be programmed into AI now so that no-one can use AI-based weapons technology to kill other humans. And if there isn't such programming then scientists working on AI are Oppenheimers:-



What we have to recognise now is that AI is already killing, the robots are killing. It is important to recognise the Vietnam factor. Vietnam was a war fought on foreign lands in my view for dubious reasons. The US establishment, particularly, suffered from internal protest and strife because of their involvement with the war - especially when US citizens came home dead. From that time technology was developed so that war could be waged without involving US troops on the ground. Decisions for war are already being taken when home lives are not being lost, decisions for war are being made using drones and smart bombs - not people. This is AI. Recognise now that AI is killing where politically NATO has difficulty justifying war with troops. We need Asimov's 3 laws now.

No more drones, no more AI-based weaponry.

As I said the problem is nationalism and racism. Because of the arbitrary "War on Terror" all people who suffer at the hands of the US are labelled terrorist and that justifies the use of AI to kill people. The 3 laws need to be applied globally so that no AI-enabled weapon can be used to kill people. If we allow one nation such as the US or NATO to define terrorism and legitimise the murder of terrorists by AI, we are already in the doomsday scenario that science fiction writers have described.

Science needs to stop being Oppenheimers. Oppenheimer did not drop the bomb on Hiroshima, but he did create the bomb for people who are hawkish enough to drop such bombs. Those hawkish people are westerners, NATO. Oppenheimers need to stop working for NATO hawks until the laws of AI robotics have been established.

Of course that will not happen because so many scientists jobs and families depend on their being ostriches.