Overcoming Section 101 Rejections for Software and AI Inventions
Computer technology patent applications often face hurdles related to “patent eligible subject matter” under 35 U.S.C. 101 (“Section 101”). Examiners sometimes wrongly allege that these important technology innovations are mere “abstract ideas.” This is rarely the case, and being able to overcome these rejections is critical for patent attorneys working in the artificial intelligence and computer technology space. In many cases, an examiner characterizes a detailed computer system at a very high level—for example, as organizing information, performing calculations, or managing a transaction—and then characterizes that general concept as an abstract idea.
At Lambert Shortell & Connaughton, we have had success overcoming 101 rejections by bringing the analysis back to the actual technology. We identify the technical problem addressed by the application, explain the particular way the claimed system solves that problem, and show where that solution appears in the claims.
The Importance of Ex Parte Desjardins
The USPTO’s precedential decision in Ex Parte Desjardins provides useful support for this approach. The case involved a method of training a machine-learning model, and the USPTO emphasized that software can provide a patent-eligible improvement to computer technology even when the improvement is defined through logical structures and processes rather than new physical hardware.
The USPTO credited benefits including reduced storage, reduced system complexity, streamlining, and preservation of performance on earlier tasks. The claims were found eligible because, considered as a whole, they improved how the machine-learning model itself operated.
That is an important point for AI and software applicants. The relevant improvement does not have to be a new physical component. It may lie in how the computer, model, database, network, or other technical system functions.
Keeping the Focus on the Actual Technology
A strong Section 101 response should do more than argue that the claims use computers. It should explain the technical shortcoming in existing systems, the specific solution provided by the application, and the claim limitations that produce that result.
The USPTO’s guidance following Desjardins instructs examiners to review the specification for a genuine technological improvement and then determine whether the claims reflect that improvement. Conclusory statements that a system is faster or more efficient may not be enough. The application should explain why the improvement occurs and how the claimed features produce it.
It is also important to resist oversimplification. In our experience, many Section 101 disputes begin when the examiner summarizes a detailed claim so broadly that the meaningful technical limitations disappear. Desjardins reinforces that claims should be considered as a whole and should not be reduced to a generalized description that ignores the claimed arrangement.
Using Examiner Interviews Effectively
Examiner interviews can be especially useful in these cases. A concise presentation often helps shift the discussion away from broad labels and toward the way the claimed system actually operates.
In one interview agenda, for example, we identified the specifically claimed system elements, and then explained how the claimed system solved Internet-specific problems.
That kind of presentation is often more effective than debating abstract terminology in isolation. It gives the examiner a concrete explanation of the technical problem, the claimed solution, and the practical improvement produced by the system.
Drafting with Section 101 in Mind
The best time to address Section 101 is before the application is filed. For software and AI inventions, the specification should clearly describe the technical problem, the particular implementation used to solve it, and any resulting improvements in performance, storage, complexity, security, processing, or system operation.
When a rejection does arise, we use the application’s own technical disclosure, the claim language, current USPTO guidance, and examiner interviews to present the invention in a practical and focused way.
At Lambert Shortell & Connaughton, this approach has been effective in helping examiners focus on what the software actually improves, rather than reducing the invention to a generalized abstract idea.