Recent Analytics, Inc. v. Fox Corp.No. 23-2437 (Fed. Cir. 2025) – On April 18, 2025, the Federal Circuit affirmed the district court’s dismissal of the case on the basis that the patent was not eligible under § 101.
background
Recentive owns four patents related to schedule optimization for television broadcast programming (U.S. Patents 11,386,367, 11,537,960, 10,911,811, and 10,958,957). Despite belonging to two different families, all four patents rely on machine learning techniques to perform their claimed methods. As described in one specification of the patent, the claimed invention includes “gradient-boosted random forests, regression, neural networks, decision trees, support vector machines, Bayesian networks; [or] other types of techniques. ”
Recentive asserted all four patents against Fox in the U.S. District Court for the District of Delaware. Fox filed a motion to dismiss the suit for failure to state a claim. Fox asserted that all of the asserted claims were invalid because they did not recite patent-eligible subject matter under 35 USC § 101. alicethe district court held that the asserted claims were unpatentable. Accordingly, the district court dismissed the case. Recently, an appeal was filed.
problem
Whether claims that simply apply established machine learning methods to a new data environment are patent eligible.
Findings and inferences
CAFC said no. Reconsidering the district court’s rejection, the Federal Circuit conducted its own review. alice analysis.
in alice In Step 1, the Federal Circuit held that the claim was directed to an abstract idea of applying common machine learning techniques. The CAFC confirmed for the first time in its records that Recentive admitted to claims that refer to the concept of broadcast scheduling, which is performed by humans and predates computers. Recentive argued that the application of machine learning to implement these concepts was the subject matter of the patent claims. However, the Federal Circuit found that the claims simply applied traditional machine learning techniques and did not make any improvements to the training model. Subsequently, Recentive argued that its claims were directed toward new applications of machine learning techniques that would speed up human activities themselves and represent technological improvements. Based on several precedents, the Federal Circuit rejected this argument.
in alice In Step 2, the Federal Circuit emphasized that Recentive’s claims overlapped with those in Step 1 and were “clearly unsuccessful.”[ed] Identifying in the claims what could “transform” the claimed abstract idea into a patent-eligible application. ”
Ultimately, the Federal Circuit recognized that machine learning is an important field and cautioned that its decision should not be interpreted as a categorical prohibition on machine learning claims. More than that-“[t]Currently, we only determine that patents that simply claim the application of general-purpose machine learning to new data environments without disclosing improvements to the applied machine learning model are not patent eligible under § 101. ”
