Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!cmcl2!yale!kthomas
From: kthomas@yale-zoo-suned..arpa (Kevin Thomas)
Newsgroups: comp.ai
Subject: Re: Who owns the output of an AI?
Message-ID: <18997@yale-celray.yale.UUCP>
Date: Mon, 16-Nov-87 10:07:14 EST
Article-I.D.: yale-cel.18997
Posted: Mon Nov 16 10:07:14 1987
Date-Received: Wed, 18-Nov-87 01:12:33 EST
References: <1778@svax.cs.cornell.edu>
Sender: root@yale.UUCP
Reply-To: kthomas@yale-zoo-suned.UUCP (Kevin Thomas)
Organization: Yale University Computer Science Dept, New Haven CT
Lines: 20
Summary: the user owns the result

Distribution:


In article <1778@svax.cs.cornell.edu> houpt@svax.cs.cornell.edu (Charles (Chuck) Houpt) writes:
>   Is this fare? Should copywrites go to the user or the programmer?
> If my AI program discovered
>a new high temperature super-conductor, shouldn't I get some profit?

The copyrights and patents should all go to the user, absent any contractual
agreements to the contrary.  This is the same debate that went on about
10-15 years ago with compilers.  Updated to the mid-80's, if I write a
program in Turbo C that Peugeot sells, should Borland be entitled to royalties?
The answer is "no, unless they say so in the sale contract, and the buyer
clearly agrees to the language in that contract".

Actually, in the case of derived products, it's worse:  If Peugeot uses a
Turbo C program to design a car, should Borland get a cut of the profits that
result from the sale of the car, in the absence of any language in the
sale contract?  I would again say "no".  Borland is free to put language
into the contract that does or does not reserve whatever rights it wants or
does not want.

/kmt