Two recently issued decisions by federal courts highlight the uncertainty around claims to software-implemented inventions after the Supreme Court decision in Alice v. CLS Bank. Both decisions relate to the patent eligibility of software that uses rules to process information and generate resultant data. In both cases, the claimed inventions use software to automate tasks that otherwise may be done by humans, with results that are apparently more accurate than prior systems. However, the court in the first case held that the claims were invalid under 35 U.S.C. § 101 for claiming unpatentable subject matter, while the court in the second case reached the opposite outcome.

The first case is Sound View Innovations v. Facebook, Inc. No. CV 16-116-RGA, 2016 WL 4535345 (D. Del. Aug. 30, 2016). The subject claims purported to improve how computers distribute information (such as news articles), by taking into account not just a user's preferences, but those of the user’s acquaintances. Specifically, the claims required using a “decision rule” to use those acquaintances’ preferences in providing information to a user. The court found the claims invalid, as directed to the abstract idea of “offering more meaningful information to an individual based on his own preferences and the preferences of a group of people with whom he is in pre-defined relationships.” Id. at *4. Notably, the court focused on the non-technicality of the claims, finding that the claimed function of selecting information on both personal and group preferences was not a “computer-specific solution,” but that the claims instead sought to “make a non‑technological improvement to the non-technological problem of providing a user with information.” Id..

The second case, McRO, Inc. v. Bandai Namco Games America, also involved claims directed to using rules to process information. Case No. 2015-1081 (Fed. Cir. Sep. 13, 2016), slip op. 9. Specifically, the claimed invention sought to use rules to help automate the selection of facial expressions for computer-generated characters. The court relied on a concession by the defendant that prior art processes were driven by “subjective determinations.” The claimed invention purported to select facial expressions more quickly and accurately, by using rules that defined how different facial expressions could be blended together based on the parts-of-speech ("phonemes") those expressions sought to depict. In deciding whether the claims recited eligible subject matter, the court stated that “the computer here is employed to perform a distinct process to automate a task previously performed by humans,” and distinguished from prior art that failed to “evaluate sub-sequences, generate transition parameters or apply transition parameters to create a final morph weight set.” Thus, the court concluded that “[i]t is the incorporation of the claimed rules, not the use of the computer, that ‘improved [the] existing technological process’ by allowing the automation of further tasks.” Thus, the court distinguished Flook, Bilski, and Alice, holding that the claims were not directed to an unpatentable abstract idea.

Despite the high-level similarities in the subject matter of these cases, the two decisions are not difficult to reconcile. The claims in Sound View were directed to the “a non-technological improvement to [a] non-technological problem.” 2016 WL 4535345, at *4. The claims in McRO were directed to a “technological improvement” to 3-D animation, which the court found to be a “technological process.” Case No. 2015-1081, slip op. 23. In sum, subject-matter eligibility for software may rest on whether the invention is viewed as sufficiently “technological”. Where an invention automates something previously done by humans, claims will more likely be found patent eligible if they include implementation details that differ from the way humans would perform the underlying task.