Sun Sep 15 16:15:16 UTC 2024: ## AI Patents Face Scrutiny: PTAB Decisions Shed Light on §112(f)

**Washington, D.C. -** As Artificial Intelligence (AI) continues to revolutionize industries, securing patents for these groundbreaking technologies is crucial. However, navigating the patent process for AI inventions can be challenging due to the complex nature of machine learning and neural networks. This complexity often leads to claim construction issues under 35 U.S.C. § 112(f), which addresses means-plus-function claim limitations and can impact how AI inventions are interpreted during the patent process and enforcement.

Recent decisions by the Patent Trial and Appeal Board (PTAB) highlight the importance of clear and detailed specifications, especially when describing the structure and function of AI algorithms within patent applications. The PTAB has consistently emphasized the need to connect functional claim language to specific structures, particularly in computer-implemented inventions.

In several recent cases, the PTAB has ruled that claims using generic terms like “algorithm,” “module,” or “resource” without providing adequate structural details in the specification, can be deemed indefinite under § 112(f). This means that the claims may lack sufficient clarity to define the scope of the invention and could be invalidated.

For example, in the case of *Ex Parte Joon Woo Son*, the PTAB found that the term “learned autonomous driving algorithm” was not sufficiently specific to define the underlying structure of the algorithm and therefore triggered § 112(f), leading to an indefiniteness rejection.

However, the PTAB also demonstrated in *Ex Parte Jing Lan* that a functional term like “processing resource” can be considered definite if the claim context and the specification clearly link it to a known structure, such as electronic circuitry.

These decisions underscore the critical need for patent practitioners to understand how § 112(f) applies to AI inventions and to adopt best practices for drafting and prosecuting AI patent applications.

**Key Takeaways for Patent Practitioners:**

* **Provide detailed descriptions:** The specification should clearly describe how the claimed functions are performed, especially for software and AI-related inventions.
* **Avoid generic terms:** When using functional terms like “algorithm,” “module,” or “resource,” ensure that the specification provides clear and detailed structural support.
* **Link to known structures:** Clearly connect functional terms to known structures, such as specific algorithms, hardware configurations, or processing steps.
* **Go beyond general-purpose computers:** For computer-implemented inventions, disclosing just a general-purpose computer without specific algorithms is insufficient under § 112(f).

By following these best practices, patent practitioners can help their clients navigate the complexities of AI patent law and ensure their inventions are protected effectively.

Read More