IP Intelligence

Home | IP Intelligence

<img class="alignright wp-image-2426 size-medium" src="https://copyrightcontentplatforms.bakerhostetlerblogs.com/wp-content/uploads/sites/21/2016/03/Computer-in-Grass-12458330-300x200.jpg" alt="Computer in Grass-12458330" width="300" height="200">New England was aflutter last week, and not just because three judges were named to hear the appeal of the National Football League against Tom Brady in the continued “deflategate” saga. The excitement continued as the media stage of the Northeast opened to Boston Dynamics, a robotics company and wholly owned subsidiary of Google, Inc., which showcased videos of its humanoid robot, Atlas.

As described by the Boston Dynamics website, Atlas (the robot) is “<a href="http://www.bostondynamics.com/robot_Atlas.html">is a high mobility, humanoid robot designed to negotiate outdoor, rough terrain</a>.” The videos released last week show Atlas maneuvering through the snow, opening doors, moving weighted boxes, and recovering from a human-like fall to the ground. The videos have garnered immediate excitement from the science fiction community, many of whom are still riding the wave of the spherical BB-8 introduced via Disney’s recent installment of Star Wars. Others have taken the time to reflect on the larger picture. For example, <a href="http://video.cnbc.com/gallery/?video=3000496516">Insider.com CEO Jason Calacanis</a> described the potential effects for society, saying “[m]anual labor is going to end in our lifetime, and in this video you can see how close we really are. It’s a huge societal issue with jobs, but it’s going to be a huge lift in terms of efficiency of companies that nobody expected.”

Wherever you stand or fall on this topic, most technologists believe that we are still many years from developing Asimovian legislation to govern our daily interactions with robots. That robots are robots and humans are humans seems clear enough. Then again, just a few weeks ago, Google received <a href="http://isearch.nhtsa.gov/files/Google%20--%20compiled%20response%20to%2012%20Nov%20%2015%20interp%20request%20--%204%20Feb%2016%20final.htm">a letter</a> from the National Highway Traffic Safety Administration (NHTSA) stating that NHTSA will interpret “driver” in the context of Google’s described motor vehicle design as referring to the self-driving system, and not to any of the vehicle occupants. The letter now appears to have been particularly timely, with reports of a February 14, 2016 accident in which a Google car collided with a public bus.  Google has released a statement saying that it bears “<a href="http://www.engadget.com/2016/02/29/google-self-driving-car-accident/">some responsibility</a>” for the accident. Although no injuries were reported, it’s not hard to imagine an autonomous car causing a more serious accident, forcing regulators and legislators to reckon with the issue of a robot’s liability.

This response letter from a government agency demonstrates a shift from prior legal interpretation, such as the California Department of Motor Vehicles <a href="https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/auto">draft proposal</a> to keep human drivers “in the loop.” The NHTSA’s letter brings up, among other issues, a question: “Will other rules or legislation begin to be interpreted in the same light?”

Last month, BakerHostetler partner Chad Rutkowski published <a href="http://www.copyrightcontentplatforms.com/2016/02/the-monkey-in-the-machine/">a post</a> that posed additional questions relating to the interpretation of works created, in whole or in part, by artificial intelligence. In addressing Mr. Rutkowski’s intriguing hypothesis, maybe the most direct path to legislative clarity relies on the country’s agency of renaissance, the NHTSA.

Other federal, state, or local agencies or judiciaries may need to follow suit and interpret rules and definitions to embrace artificial intelligence, including humanoid robots. As an example, the California judiciary applies a pure comparative negligence standard of liability, adopted in <em>Li v. Yellow Cab Co</em>., 13 Cal. 3d 804 (1975). Under this standard, each defendant is liable only for his or her percentage of fault. In determining the assignment of fault, factors such as violation of a statute or analysis of the facts under a “reasonable person” standard may be considered. Will courts need to develop and apply a “reasonable computer” standard?

Similar logic may carry over to other bodies of law, such as copyright law. <a href="http://www.kurzweilcyberart.com/">Computer programs</a> are training through various machine learning techniques to develop a truly creative artistic expression. Just as an apprentice studies art in order to create art, a computer can study art in order to create its own original art. Why then can an apprentice become an author of an original work, but a toiling machine must be forced to consign its works to the public domain? Maybe the U.S. Copyright Office or the judiciary is now warranted in interpreting “author” as including artificial intelligence.

Or maybe it’s just time to dust off your copy of Asimov’s “I, Robot.”

The Reasonable Computer