3 Laws of AI-tics Proposed

[BillTre]

Put here rather than in a computer forum because it is at the interface of computing and society.

Many are familiar with Asimov's 3 Laws of Robotics:

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm
2. A robot must obey the orders given it by human beings, except when such orders would conflict with the previous law
3. A robot must protect its own existence as long as such protection does not conflict with the previous two laws.

This article in the NY Times proposes analogous laws for modern day AI:

1. An A.I. system must be subject to the full gamut of laws that apply to its human operator.
2. An A.I. system must clearly disclose that it is not human.
3. An A.I. system cannot retain or disclose confidential information without explicit approval from the source of that information.

The reasons for these (in the article) seem well thought out to me.

However, how regulation could guarantee implementation is not clear to me
I have the same problem with the robot laws, since the stories seemed to have benign companies, while today I perceive rather predatory companies.

 

[Choppy]

This is a really interesting topic for discussion.

I might suggest a slight difference in the wording of rule 1. There's no reason an AI system must have an operator. Isn't the idea that an AI system could be "autonomous?"

There are also a questions of enforcement. Do you penalize the operator/designer/owner? Does violation mean immediate shut down? If so, what happens when f the system is running something vital to human survival? Who validates the code or operation? How?