Patent Infringement

3. When the entities are carrying out a joint enterprise

Slip Op. at 6-7. Under this framework, merely encouraging or teaching an entity to carry out the steps of a patented method will not result in direct infringement. Slip Op. at 7.

U.S. patent law defines three types of patent infringement: 1) direct infringement, 2) contributory infringement, and 3) active inducement of infringement.  Direct infringement is a strict liability tort, so one can directly infringe a patent without knowing it existed.  Contributory infringement and active inducement of infringement require knowledge of the patent.  In general, one commits contributory infringement by supplying part of a patented invention that has no substantial non-infringing uses.  One commits active inducement of infringement by taking steps to cause another party to infringe a patent.  For example, one can be held liable for inducing infringement by supplying a component with non-infringing uses but teaching or encouraging third parties to use it in an infringing manner.  As discussed further below, the Supreme Court confirmed last year that a party can only be held liable for inducing infringement or  if someone directlyinfringed the patent. 

In the law of torts, multiple parties can often held jointly liable for causing injury to another.  Akamai argued that these common law, joint tortfeasor principles should determine whether multiple parties may be jointly liable for direct infringement under the Patent Statute.  The Federal Circuit disagreed and held that Congress codified the limits on joint tortfeasor liability when it enacted specific rules for contributory infringement and active inducement of infringement, indicating an intent not to apply the general law of joint tortfeasors to direct infringement.  Slip Op. at 11-12.  The Federal Circuit stated that “Congress carefully enacted sections (b) [active inducement of infringement] and (c) [contributory infringement] to expressly define the only ways in which individuals not completing an infringing act under § 271(a) [direct infringement] could nevertheless be liable.”  Slip Op. at 11-12.  

The patent-in-suit in Akamai concerns methods for providing “content delivery networks” (“CDNs”), which provide a faster and more efficient way of delivering content such as web pages at global locations across the Internet. The accused infringer (Limelight) performed all of the steps of the claimed method, except for one which was carried out by Limelight’s customers.  Limelight gave its customers instructions for carrying out the final step.

In an earlier decision in 2012, the Federal Circuit held that Limelight could be liable for inducing infringementof a method claim (not direct infringement) if, collectively, Limelight and its customers carried out all steps of the patented method, even if no one could be held liable for direct infringement.   We discussed that decision in a September 2012 blog post.

In June 2014 the U.S. Supreme Court reversed the Federal Circuit, holding that inducing infringement can only occur when some party is liable for direct infringement. In other words, one cannot induce an act of infringement that never actually occurred. We discussed the Supreme Court decision in our June 2014 blog post.   Notably, the Supreme Court expressly declined to address whether the Federal Circuit’s BMC and Muniauction opinions were correct, and thus, did not discuss whether and under what circumstances a method claim could be directly infringed by multiple actors.  It would, therefore, be unsurprising if the Supreme Court weighs in on these issues this time around.

Judge Moore dissented in this most recent Akamai decision.  She disputed the majority’s contention that the “single entity rule” as articulated in BMC and Muniauction was a longstanding, established rule for deciding questions of direct infringement.  Dissent at 7.

The question of how to handle divided infringement of method claims has become increasingly important with the advent of the Internet.  Inventions such as the one at issue in Limelight often involve the interaction of different components (e.g., servers) across the Internet.  In some cases, patent claims can be drafted to ensure that there is only one entity involved in practicing the claimed method, which should avoid the divided infringement problem. However, in other cases, it may not be clear at the time of the invention how different actors could carry out different aspects of a method.  In addition, claim drafting rules may foreclose effective ways of defining what is novel and non-obvious based on the acts of one entity alone.  In such circumstances, the patent holder may effectively be without recourse despite the fact that others are using and profiting from the invention.  As the Federal Circuit explains in Akamai, the divided infringement problem is not typically an issue with apparatus claims because with apparatus claims, there is always a single entity that is responsible for making, using, selling, or importing the apparatus.  Slip Op. at 18.