The Alice Test Automated: A Fast, Cost Effective, Up-to-Date Solution for a Fluid, Ambiguous Law

Determining patent eligibility of a claimed invention under § 101 of the Patent Act can turn into a complicated, daunting task. The Supreme Court of the United States has introduced a two-step framework named the Alice test to assess patent eligibility of a claim. The main challenge is that the judicial branch has not provided an exact definition or a guideline for the Alice test. Instead, the court says look at the prior litigated cases and how the court has applied the Alice test and use them as guidance on how to determine the patent eligibility of your claim.

The Alice test has been a great problem for IP owners and patent law practitioners. Stacks presents Selena. Selena is a data-driven system that has combined all of the prior litigated claimed inventions under the Alice test in a systematic way, and applies the collective knowledge of these prior cases to a new claim or invention to predict and determine its eligibility under the Alice test.

The Alice Test

The patent laws are designed to strike a right balance between promoting innovation by awarding exclusive rights to the inventors, but at the same time not to inhibit future innovation by granting a monopoly over certain fundamental principles. Section 101 of the Patent Act defines patentable inventions as "any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof."

The Supreme Court of the United States has described "abstract ideas, laws of nature, and natural phenomena" as the fundamental principles and have excluded them from the patentable inventions. For this end, the Supreme Court has introduced a two-step framework commonly referred as the Alice test to assess and determine patent eligibility of a claimed invention. A diagram of the Alice test is shown in Fig. 1.

At step 1 of the Alice test, determine if the claimed invention is directed to "abstract ideas, laws of nature, and natural phenomena." If the claim is not directed to one of these fundamental principles, then the claimed invention is patentable under § 101. Otherwise, at step 2 determine if the claim involves an inventive step that transforms the ineligible fundamental principle to a patent eligible application.

¹ The 101 eligibility assessment involves an initial test to determine if the claimed invention fits into one of the four statutory categories, namely "process, machine, manufacture, or composition of matter." If not, the claimed invention is not a patentable invention under § 101. However, such scenarios are not common, and when they happen, they are easy to detect and address. The main challenge and concern over § 101 eligibility relates to the judiciary exceptions, embodied by the Alice test. The solution reported here is focused on the Alice test.

This seemingly simple analysis is, however, difficult to perform. The main challenge is that in the United States patent law there is no definitive or exact rule on what an abstract idea is and what is considered as an inventive step. For example, the Supreme Court of the United States in Alice Corp. v. CLS Bank Int'l says that:

We need not labor to delimit the precise contours of the 'abstract ideas' category in this case.

Similarly, the United States Court of Appeals for the Federal Circuit in Amdocs (Israel) Ltd. v. Openet Telecom, Inc. explains that:

However, a search for a single test or definition in the decided cases concerning § 101 from this court, and indeed from the Supreme Court, reveals that at present there is no such single, succinct, usable definition or test.

This absence of a definitive, exact rule is not by mistake or a fault, rather it is intentional and by design. For example, the United States Court of Appeals for the Federal Circuit in the same published opinion describes this point:

The problem with articulating a single, universal definition of `abstract idea' is that it is difficult to fashion a workable definition to be applied to as-yet-unknown cases with as-yet-unknown inventions. That is not for want of trying; to the extent the efforts so far have been unsuccessful it is because they often end up using alternative but equally abstract terms or are overly narrow.

² This court in the same decision describes a mechanism to determine patent eligibility of a claim in the absence of an exact definition:

Instead of a definition, then, the decisional mechanism courts now apply is to examine earlier cases in which a similar or parallel descriptive nature can be seen — what prior cases were about, and which way they were decided. See, e.g., Elec. Power Grp., 830 F.3d at 1353-54. That is the classic common law methodology for creating law when a single governing definitional context is not available.

Basically, the court says that one must look at how prior cases were decided and use them as the guidance to determine patent eligibility of new claimed inventions.

Why the Alice Test is So Important

Software patents are greatly impacted by the Alice test. Almost any software-based invention faces scrutiny over if it is abstract and whether it includes an inventive step or not. About 70% of the new patents are software-based or at least include a software component, and this ratio is increasing. As a result, the Alice test is affecting a growing majority of the patents in the US.

Some of biotechnology and pharmaceutical inventions may face the question of if they are a natural phenomenon or a law of nature, and therefore they could be subjected to the Alice test as well.

Furthermore, the judicial interpretation of the Alice test has been fluid and changing over time. A new case law can turn a valid existing patent into an invalid patent overnight and vice versa. Therefore, the patent eligibility of a claimed invention under the Alice test is not a question that can be answered for once and settled forever, rather it requires recurring evaluations according to the latest interpretation of the Alice test in the new litigated cases.

Current Solution to the Alice Test

Currently, to determine subject matter eligibility of an invention under the Alice test, the claimed invention needs to be examined by an expert patent lawyer. The patent lawyer usually performs a legal research to gather Alice-related case files in which similar technologies and inventions are litigated. And following these case laws, applies a similar logic to the invention under study to determine its patent eligibility under the Alice test.

There is no specialized legal research tool for the Alice test, and general-purpose legal research tools are commonly used to locate the Alice cases. Some firms even have prepared their own spreadsheet of existing Alice cases (with no analytics).

Of course, the process of having each invention or a claim examined by a patent lawyer is a costly and time-consuming process. Furthermore, this may not be a viable solution in many situations such as:

• IP owners, in-house counsels, and other stakeholders who usually do not have enough budget for frequent and expensive outsourcing to assess their large patent portfolio.
• IP marketplaces where a huge number of patents are traded, and therefore there is neither time nor means to perform such manual assessment of validity of each and every patent.
• Investment and financial institutions who receive many applications from companies with a large patent portfolio, and therefore they cannot review the IP strength of each application at reasonable time and cost.
• Innovation and R&D centers who need to assess patent eligibility of the ideas before hiring expensive lawyers to file a patent or investing more time and resources into those ideas.

Currently, IP owners and other stakeholders have two options: 1) the costly and time-consuming option of having IPs and inventions manually examined by a patent lawyer, and 2) neglecting the patent eligibility of the IPs and the inventions, therefore taking the risk of higher costs down the road. There is a need for a third option; an automatic tool that can perform patent eligibility of the inventions and patents under the Alice test at scale, at a reasonable cost, and quickly. Such a tool alone can address many of the needs listed above, or at least it can reduce the number of or the amount of outsourcing to the expert lawyer.

Machine Learning and the Alice Test

Machine learning is a data driven approach to extract and to learn underlying patterns in the data and to use them to predict the outcomes of the new cases. Machine learning methods are extremely useful when there is no exact, or easy-to-use rule present, however, there exists a large body of prior data to learn from and to design a predictive model.

There is an intriguing parallel between how the court prescribes to apply the Alice test - using the prior cases as a guidance for the new cases and what machine learning provides as a solution, therefore making machine learning an ideal method to automate and to implement the Alice test.