Posted on

Invented by machine?

When Robert Plotkin, our IP lawyer, came to my office and tried to explain this to me a few years ago, I got the gist but still found it hard to explain the concept to someone else. Now the book he came to get my advice about is a reality, and this review of The Genie in the Machine in Library Journal provides greater clarity than I could manage:

Plotkin, Robert. The Genie in the Machine: How Computer-Automated Inventing Is Revolutionizing Law & Business. Stanford Law & Politics: Stanford Univ. 2009. c.288p. index. ISBN 978-0-8047-5699-0. $29.95. BUS

There is little argument that invention spurs innovation, competition, and economic growth. With technology today, however, inventors can simply input a problem (a “wish”) into a program and have the computer (a “genie”) generate, or “invent,” the ultimate solution. Who or what, then, is the true inventor of the final product? Plotkin, an intellectual property attorney, tackles this intriguing question by stating that patent law today does not lend itself to such broad interpretation. Further, the author convincingly illustrates an urgent need to reform current law so that it is neither too strong nor too weak in order to protect the future rights of inventors, businesses, and consumers. VERDICT Plotkin posits that “Computer Automated Inventing” or “Artificial Invention Technology” does not replace the human mind; rather, it augments and partners with its human counterpart to build a better mousetrap, whatever that might be. From toothbrushes to auto assembly, the author uses easy-to-understand analogies that most lay readers will understand. Recommended for committed readers in business, computer science, or law.—Judy Brink-Drescher, Dowling Coll., Oakdale, NY

“Invention” is obviously another topic for a publication I’ll soon get underway, to update and expand from the Berkshire Encyclopedia of Human-Computer Interaction. Think of all the developments: micro-blogging, mash-ups, and even computerized customer service!

3 thoughts on “Invented by machine?

  1. Invented by machine…how about written by machine. Who owns the copyright? Take for example machine generated summaries of copyrighted articles. A person does it and that person owns the copyright with no obligation to the author of the full length article. Now, what if two people own the same software…and apply it to the same article. Who owns the copyright when two different people applied the same software to the same article? What happens if the original author also buys that software and applies it to their own article?

  2. Great question! I will give some possible answers to it, although admittedly I am much more conversant in the implications of automation for patent law than for copyright law.

    U.S. copyright law, unlike patent law, requires some degree of “creativity” to go into creation of a work in order for it to be copyrightable. Therefore, arguably computer-written works are not copyrightable if they are created without any creativity. However, this just begs the question of what we mean by “creativity.” (Read Margaret Boden’s “The Creative Mind: Myths and Mechanisms” for an extended treatment of this.)

    Now let’s assume that a computer-written work is copyrightable to the person who uses the software to create the work. If two people apply the same software to the same article, I would think that the first person to do so owns the copyright in the computer-generated summary. The summary created by the second person no longer satisfies copyright law’s “originality” requirement, since it is not the first such summary ever created. If this seems like a bizarre outcome, it is probably a good reason for not concluding that the first person should have been granted a copyright in the summary in the first place.

    Even if the first person is granted a copyright in the summary because he or she happened to have been the first person to create the summary, subsequent people who use the same software to create the same summary do not “infringe” (violate) the first person’s copyright, because they did not create the subsequent summaries by copying the first summary (they created their summaries from the original article), and copying is required for copyright infringement. (Note that in patent law, copying is not required for infringement; if I independently create a patented invention and sell it, I am still infringing the existing patent, even though I did not copy it.)

    Other people have given the topic of copyright protection for computer-generated works much more thought than I have. If you are willing to read through long academic legal articles, two of the best written on this topic are:

    Pamela Samuelson, “Allocating Ownership Rights in Computer-Generated Works,” 47 U. Pitt. L. Rev. 1185 (1986) (

    Ralph Clifford, “Intellectual Property in the Era of the Creative Computer Program: Will the True Creator Please Stand Up?,” 71 Tulane L. Rev. 1675 (1997)

  3. Thanks for the response! Perhaps we should grant the copyright to the software creator since that is where the creativity was real!

    Your point about copying required for copyright infringement and not for patents is exactly why I am vehemently opposed to software patents. Software should be copyrightable…you are just writing in a different language. Imagine the burden of knowing whether some lines of code are somehow the same in the billions of lines that are written. More to the point, if there were software patents in the early days and everything done the first time with a computer program was patented (think Amazon’s one-click purchase patent), like: sorting, requesting input, printing, converting data types, formatting, click ‘ok’ to proceed….it would have been too risky to write software, leaving it to IBM, stifling competition and creativity instead of encouraging it.
    Thanks for the references.

Leave a Reply

Your email address will not be published. Required fields are marked *