Who owns photos taken by monkeys today? Tomorrow, who will own photos taken by autonomous drones, or written works produced by machine learning algorithms or artificial intelligence?
Copyright lawyers have been having a field day over the “monkey selfies”, an iconic series of photos which were taken by a monkey in Indonesia with photographer David Slater's camera. Slater painstakingly set up the camera and waited patiently, and the curious monkey picked up the camera and snapped some selfies:
Slater then got into a dispute with the Wikimedia Foundation for its hosting and wide distribution of the photos. Slater claimed he owned the copyright to the photos, and Wikimedia claimed they were not copyrightable because they were taken by an animal, instead of a human—and only humans can create copyrightable works.
Last summer, the Copyright Office released a draft of the new Third Edition of its Compendium of U.S. Copyright Office Practices. The Compendium is gargantuan, weighing in at over 1,200 pages.
The Third Edition addressed the monkey selfie problem head-on, concluding that the photo is not copyrightable at all. The rule behind this conclusion is that copyright attaches only to works by humans. Section 306 of the Compendium states: “The U.S. Copyright Office will register an original work of authorship, provided that the work was created by a human being.” Section 313.2, titled “Works That Lack Human Authorship”, hammers the point home, specifically ruling out the monkey selfie, along with several other examples:
- A photograph taken by a monkey.
- A mural painted by an elephant.
- A claim based on the appearance of actual animal skin.
- A claim based on driftwood that has been shaped and smoothed by the ocean.
- A claim based on cut marks, defects, and other qualities found in natural stone.
- An application for a song naming the Holy Spirit as the author of the work.
The Compendium cites Burrow-Giles Lithographic Co. v. Sarony, an 1884 Supreme Court case about a posed photo portrait of Oscar Wilde, for the proposition that "To qualify as a work of 'authorship' a work must be created by a human being."
It's no secret that the Internet loves cat videos.
In 2007, Jürgen Perthold, a German engineer living in the US, created the CatCam, a compact digital camera for his cat, Mr. Lee, to wear on his collar. Mr. Lee became famous on the Internet for a time, due in part to the quirky photos and videos from the CatCam.
Sadly, this legal development appears to divest Mr. Lee's owner of the copyrights in the CatCam photos. Like the monkey selfie, Mr. Lee was out on his own with the CatCam. Mr. Perthold had no knowledge of what photos were being captured until he downloaded them from the device. Catnip Technologies, the company now behind the CatCam, claims broad copyright rights, however:
All material (text, images, video, manuals, etc.) shown on our pages is copyright protected. Some pages may contain material from other copyright owners which are not explicitly listed. Before you use material from our pages please ask for permission. If protected material from other owners is affected we will forward your request to the original owner.
Today, nearly a decade later, mobile cameras are ubiquitous. GoPro even sells the Fetch dog mount. If you give a dog a GoPro, is the resulting video public domain? It would seem so. The same would be true of this video, at least after the monkey snatches the tourist's GoPro and runs off with it.
What the Compendium addresses less directly, however, is whether works created by machines are copyrightable. It is well settled that where a human uses and directs a machine, there is a copyright right and it belongs to the human. But when does that stop being true? In today's world of increasing machine intelligence, there is much less human direction.
Take, for example, photos captured by a camera attached to a high-altitude weather balloon. No human pressed the shutter button, for sure. Presumably a human set the camera to snap photos at time intervals. Is that camera and timer setup enough to endow a human with a copyright?
Taking it to the next step, what if an algorithm decided when to take the photos? Does the fact that the algorithm was programmed by a human give that programmer the copyright? Is that fair—do we want programmers (or the companies that employ them) owning the copyrights to photos taken by products they sell into the marketplace?
Going one step further, what if a machine learning algorithm decided to take the photos? The programmer is another step (or many more) removed from the photo taking.
There is already one category of fantastic robot photos: those taken by spacecraft like NASA's Mars landers. So far, these photos are in the public domain because works by government agencies are outside the realm of copyright. When NASA lands a rover on Mars, the rover's photos are in the public domain because the government took them (at least under United States law).
But governments are losing their monopoly on space photos. Take SpaceX and its ambitions to land its spacecraft on Mars. In a fully private space mission, when an uncrewed Red Dragon lands on Mars, the photos it takes won't similarly go into the public domain—which may bring the human contribution issue front and center once again.
The question all these scenarios raise is how much authorship a human has to contribute for a work to be copyrightable. Today, there is no clear answer, but our technology is rapidly moving to the point where we will need one.
It may be that, as a society, we might want to move away from the human authorship requirement. Instead, we may want the human (or corporate) owner of a machine to own the copyright in its works. This is how it works, by analogy, in other areas of law. If my drone shoots you, I am liable because I own the drone. If I own the drone's liability, why not its photos?
Originally posted in January 2015.
This entry was posted on Friday, April 17, 2015 and is filed under Trademarks Copyrights and Trade Secrets, Internet Law News.