Recentive Analytics v. Fox—In Precedential Case of First Impression, Federal Circuit Holds Machine Learning Patents That Do Not Improve the Technology Ineligible under Section 101

On April 18, the Federal Circuit issued a precedential decision in Recentive Analytics, Inc. v. Fox Corp., invalidating under Section 101 of the Patent Act and the two-step Alice test patents using machine learning models. The Federal Circuit acknowledged that “[m]achine learning is a burgeoning and increasingly important field and may lead to patent-eligible improvements in technology.” Nevertheless, the Federal Circuit held that “patents that do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied, are patent ineligible under § 101.” Although some district courts had previously found claims on machine learning methods patent-ineligible, the Federal Circuit recognized that this case presented “a question of first impression” under its precedent.

As background, Recentive Analytics sued Fox Corp. and certain affiliates alleging infringement of four patents that purportedly addressed problems in the entertainment and broadcasting industries. The patents fell into two families. The first focused on iteratively training a machine learning model to identify relationships in collected data sets to generate an optimized schedule for live events. The second family focused on training a machine learning model to generate and update optimized network maps, which determine which programs are broadcast in different geographic markets at different times. Fox moved to dismiss, arguing that the claims were ineligible under Section 101. The district court granted the motion, and the Federal Circuit affirmed.

Under Alice step one, the Federal Circuit held that the patents were “directed to the abstract idea of using a generic machine learning technique in a particular environment.” Specifically, the patents recited “the use of generic machine learning technology in carrying out the claimed methods for generating event schedules and network maps.” The Federal Circuit cited Recentive’s admissions that the patents were “not claiming machine learning itself” and did “not claim a specific method for” improving any underlying machine learning algorithms. Citing precedent on other computer-implemented claims, the Federal Circuit rejected Recentive’s argument that applying an existing machine learning technique to a new field or task is patent-eligible, even assuming that it can “perform a task previously undertaken by humans with greater speed and efficiency than could previously be achieved.” In so holding, the Federal Circuit reaffirmed that the inquiry under Alice step one for computer-implemented inventions turns on whether the claims focus on a “specific asserted improvement in computer capabilities” or, instead, “on a process that qualifies as an abstract idea for which computers are invoked merely as a tool.”

Under Alice step two, the Federal Circuit rejected Recentive’s argument that “using machine learning to dynamically generate optimized maps and schedules based on real-time data and update them based on changing conditions” is an inventive concept. Again relying on prior case law, the Federal Circuit held that “this is no more than claiming the abstract idea itself,” which is legally insufficient. Under both Alice steps, the Federal Circuit extended its existing Section 101 precedent on other methods of collecting, processing, and displaying information—confirming that the same legal inquiries apply to patents claiming the use of machine learning.

Importantly, the Federal Circuit did not rule out the possibility that other machine learning claims could be found patent-eligible. The Federal Circuit recognized that machine learning techniques “may lead to patent-eligible improvements in technology” and emphasized that the claims on appeal did not recite specific “improvements to the machine learning models to be applied.”