Into the Future: Responding to Patent Office Actions in the AI and Simulation Modeling Arts

Group Art Unit 2128 deals with some of the most exciting and cutting-edge technology in the Patent Office, artificial intelligence and simulation modeling. However, patent examiners sometimes reduce these inventions to mere algorithms when making rejections under § 101. This reduction allows examiners to dismiss the technological improvement provided by the invention as nothing more than a mathematical concept that can be performed in the human mind. The “mental process/mathematical concept” rejection has always been a legal fiction and this can make the rejection difficult to overcome. After all, an examiner can argue that any mathematical concept could theoretically be performed in the human mind, even if no one has ever done so.

Here are three inspiring office actions that show how you can respond to a “mental steps/mathematical concept” rejection in the AI and simulation modeling art areas.

Lay the Groundwork with Your Specification

Since almost any application in this technology class is vulnerable to a rejection under § 101, you should draft your specification with a potential § 101 rejection in mind. Specifically, you can include ammunition in your specification that will ground the arguments in your office action response. This will provide credibility to your arguments, so the examiner knows you are not creating them out of thin air.

This strategy was used with impressive results by Samir A. Bhavsar from Baker Botts LLP in U.S. Patent No. 10,915,342. This patent was allowed after only a single office action response. This speedy prosecution was helped by an examiner interview that was conducted prior to filing the office action response.

The invention is a computer system for comparing data sets that cannot be quantified in a simple way. For example, the invention could be used for recognizing a face based on a reference photo or recognizing a voice from a voice sample.

The examiner rejected the application under § 101 on the ground that the claims merely recited a series of steps that could be performed in the human mind. In response to this mental steps rejection, the applicant argued that the invention improved technology through a practical application of the algorithm recited in the claims.

This argument is nothing new. This is often the most effective argument in response to a mental steps rejection. But in this office action response, the applicant diligently pointed out several places in the specification that described both the purpose of the invention and the practical implementation of the invention.

[T]he claims recite elements that are integrated into a practical application, and are therefore not directed to an abstract idea. Although other practical applications exist, Applicant will focus on the practical application directed to cloud-based computing, as explicitly recited in amended Claims 1, 8, and 14.

For example, the specification describes, “Another technical advantage of a correlithm object processing system is their ability to be adapted to implement a cloud based architecture that allows correlithm objects to be processed remotely as cloud services. For example, a device may send correlithm objects to a cloud based correlithm object processing system to offload the resources used for processing the correlithm objects. In this example, the device is able to receive processed correlithm objects from the cloud based correlithm object processing system without having to consume the device’s processing resources. The cloud based correlithm object processing system allows devices to utilize the benefits of correlithm objects (e.g. noise immunity and information security) while offloading the computing resources.” See Specification at p. 8, ll. 19-28 (emphasis added).

The specification set up the argument so well that the legal argument takes only a few lines in over two pages of the response. The remaining space is occupied by extensive quotes directly from the specification to support the legal argument. The takeaway from this case is to write your specification to include facts about the invention, how it is used, and how it improves technology.

Claim Steps that Are More Than Mathematical Operations

Another approach was used to great effect by Steven Fischman at Scully, Scott, Murphy & Presser, P.C. in U.S. Patent No. 10,885,241. This invention is a simulation system that uses a correction operator to correct for error between observed data and model data.

This patent was originally rejected under § 101 as a mathematical concept. In this case, the applicant overcame the rejection by amending the claims to specifically recite steps that were not mathematical operations. For example, the amended claims recited “appending … the correction operator to the simulation operator to produce a supplemented operator without modifying the simulation operator.”

By amending the claims in this way, the applicant did not surrender any claim scope. But the applicant did gain the argument that the claims were “more than” a mathematical concept. Specifically, the applicant argued that the invention integrated a mathematical algorithm with non-mathematical data processing steps and that:

As a result of such integration, the claimed processor is enhanced with functionalities to generate “supplemented output data by applying the simulation model with the supplemented operator on a set of inputs”. The practical application integrated with the elements in the pending claims addresses challenges faced by simulations systems, and provides the simulation systems with ‘relatively cost effective approach in terms of development time and computation, since current practices involve significant modification of core simulation code, which may be time consuming from a development perspective, and may significantly slowdown the implementation due to increased computational complexity of the more comprehensive formulation”. See paragraph 52 of the originally filed specification.

This response weaved together two important concepts for overcoming a rejection under § 101:

  • A real-world application for the invention
  • A technological improvement that results from the practical application

The takeaway from this response is that amending the claims to recite steps that are non-computational provides a path to arguing that the invention is practical, rather than abstract.

Argue the Limitations of the Human Mind

Another approach to overcoming a rejection under § 101 comes from U.S. Patent No. 10,761,494. This patent is directed to a simulation method that predicts the value of a variable if a particular social policy is enacted. The application was originally rejected under § 101 as an abstract idea.

In response to the rejection, Gene M.Garner II from Staas & Halsey LLP persuasively argued that the method as claimed was more than just the mental concept of estimating the value of a variable in response to a social policy. Instead, the applicant focused on the steps of the method that could not be performed by the human mind. This allowed the applicant to argue that even if the idea of estimating the value of a variable could be held in the human mind, the process recited in the claims could not be carried out by the human mind.

Thus, claims 1-9 do not recite a mathematical concept that can be practically performed in a human mind because the claims do not cover performance in the human mind but for the recitation of generic computer components. For example, at least the claimed feature of “transmitting, to the terminal via the network, a simulation result including the predicted value of the objective variable in response to introduction of the policy, to display on the terminal a display screen representing a difference between the predicted value of the objective variable in the absence of introduction of the policy and the predicted value of the objective variable in response to introduction of the policy, and enable the user to quantitatively recognize the effect of introducing the policy” requires the processor’s action that cannot be practically performed in a human mind (see, e.g., 2019 PEG, footnote 14; and USPTO Example 37, claim 2, Example 38, and Example 39 issued January 7, 2019).

The office action response did not stop there. Once the applicant highlighted the steps of the method that required a processor because they could not be performed in a human mind, the applicant pointed out that these steps provided a technological improvement. The applicant wove quotes from the MPEP directing examiners to look at the specific requirements of the invention rather than oversimplifying every process into mental steps/mathematical concepts.

One takeaway from this office action response is that you should always include steps that cannot be practically performed in the human mind in your claims. The other takeaway is that you should argue that these non-mental steps improve the technological state of the art.

Overcoming Rejections Under § 101 in Art Unit 2128

These office action responses are not the only ways to overcome a § 101 rejection in the AI and simulation model domain. But they do provide some inspiration for practitioners in these fields. Even though this domain is heavily dependent on algorithms, there are ways to show that a method or system can be patent-eligible.

Use the specification to identify technical improvements and practical applications. And be sure to claim steps that cannot be performed in the human mind and relate to non-computational data handling.

IP Toolworks Demo

Disclaimer – “The statements and views expressed in this posting are my own and do not reflect those of my law firm, are intended for general informational purposes only, and do not constitute legal advice or a legal opinion.”

No comment yet, add your voice below!

Add a Comment

Your email address will not be published. Required fields are marked *