Any AI-related patent application filed in the Philippines must be scrutinized to ensure that the claimed invention does not contravene public order or established standards of morality. This foundational principle has gained renewed significance as the Intellectual Property Office of the Philippines, commonly referred to as IPOPHL, introduced specialized AI Examination Guidelines in July 2025 to address the surge in autonomous and semi-autonomous innovation. While the Intellectual Property Code of the Philippines traditionally focused on protecting the fruits of human labor, the rapid integration of machine learning and generative models has forced a reevaluation of what it means to invent in the modern era. Today, patent examiners are no longer just looking at the novelty of a mechanical process or a chemical formula; they are navigating the complex intersection of code-driven automation and human creative intent. This regulatory evolution reflects a global trend where patent offices must balance the desire to foster technological progress with the need to maintain the legal integrity of the patent system, which has always been anchored in the recognition of human genius.
The landscape of Philippine intellectual property has shifted dramatically since the beginning of 2026, as legal practitioners and innovators grapple with the practical implications of these new standards. The definition of an inventor, once considered a settled matter of biological fact, is now a point of sophisticated legal debate. Under Republic Act No. 8293, an invention must provide a technical solution to a problem in any field of human activity to be patentable. However, as AI systems begin to identify patterns and suggest solutions that no human had previously considered, the line between the tool and the creator becomes blurred. The 2025 guidelines were designed to provide clarity, yet they also highlighted the inherent tension in a legal system that was built for the industrial age but must now function in the age of artificial intelligence. As a result, the current environment requires a meticulous approach to patent drafting, where the role of the human operator must be documented with precision to ensure that the resulting intellectual property remains enforceable and secure under the domestic framework.
1. The Role: Defining Human Inventorship in the Machine Age
The core of Philippine patent law remains rooted in the concept that only natural persons can be legally recognized as inventors. This stance is not merely a bureaucratic preference but is deeply embedded in the Civil Code of the Philippines, which distinguishes between natural persons and juridical entities. Because an AI system lacks a biological existence, it cannot possess “juridical capacity,” which is the fitness to be the subject of legal relations, nor can it exercise the “capacity to act,” which is the power to perform acts with legal effect. Consequently, even the most advanced generative AI, capable of drafting complex chemical structures or optimizing aerodynamic designs, cannot be named as a primary inventor or a co-inventor on a patent application. The legal framework necessitates a human anchor—a person who provides the conceptual spark, the specific guiding prompts, or the ultimate validation of the AI’s output. Without this human connection, an invention is effectively “authorless” in the eyes of the law, potentially falling into the public domain where it cannot be commercialized or protected against competitors.
Furthermore, this human-centric requirement serves as a safeguard against the dilution of patent quality and the potential monopolization of entire fields by those who control the most powerful computing resources. If AI were granted the status of an inventor, the speed at which machines can generate “novel” iterations could lead to an overwhelming influx of patent filings that stifle rather than encourage human competition. The IPOPHL has maintained that while AI is an increasingly vital tool that augments the creative faculties of designers and makers, it remains just that—a tool. Just as a microscope or a sophisticated piece of laboratory software does not claim ownership over a discovery, the AI system is viewed as an extension of the human mind. To secure a patent in 2026, the applicant must demonstrate that a human was the one who defined the problem, selected the relevant parameters, and made the critical decisions that led to the final technical solution. This ensures that the patent system continues to reward human ingenuity and investment, rather than simply rewarding the ownership of high-end processing power.
2. The Distinction: Natural Persons Versus Juridical Entities
In the context of the Philippine legal system, the distinction between different types of “persons” is vital for understanding who can own a patent versus who can be credited with its creation. While the right to a patent belongs to the inventor, the law allows for the assignment of this right to other parties, including corporations and government agencies. These entities are known as juridical persons, and they can apply for and hold patents in their own names. However, the initial spark of invention must always be traced back to a natural person—a human being. This creates a two-tier system where the “inventor” is always human, while the “applicant” or “patentee” can be a multi-billion-peso corporation. AI systems do not fit into either of these categories. They are not natural persons because they lack human life, and they are not juridical persons because they are not recognized as legal entities created by law, such as a partnership or a corporation. Therefore, AI lacks the standing to sue for infringement, to enter into licensing agreements, or to be held liable for violating the rights of others.
This legal reality creates a significant barrier for those attempting to push for “AI-only” patent applications. Without the ability to hold property rights, an AI system cannot fulfill the requirements of a patentee, and without the capacity to be sued, it cannot participate in the adversarial process that defines intellectual property enforcement. The IPOPHL has emphasized that the 2022 Revised Patent IRR and subsequent 2025 guidelines reinforce this position by requiring clear identification of the human inventors. This approach aligns with global precedents, such as the widely discussed Thaler cases in the United States and the United Kingdom, where courts consistently ruled that AI-powered machines like DABUS could not be granted inventor status. By maintaining this boundary, the Philippines ensures that its intellectual property regime remains compatible with international treaties and standards, providing a stable and predictable environment for both local and foreign investors who seek to protect their technology in the Southeast Asian market.
3. The Procedure: Initial Steps of AI Patent Evaluation
To navigate the complexities of AI-assisted inventions, the Intellectual Property Office of the Philippines has implemented a specialized five-step claim analysis to determine the eligibility of a filing. The first step involves confirming that the invention fits within the legal categories defined under Section 21 of the IP Code. This means the examiner must verify that the proposal is a product, a process, or an improvement of either, and that it addresses a specific technical problem. In the current era, many AI applications are dismissed at this early stage if they are presented merely as mathematical models or abstract theories without a clear industrial application. The challenge for innovators is to describe their AI-driven solution in terms of its physical or digital impact on a specific field, such as telecommunications, healthcare, or agriculture. By grounding the AI’s functions in these established legal categories, applicants can move past the initial hurdle of subject-matter eligibility.
The second step of the evaluation process requires a deep dive into the specific reach and technical nature of the patent claims. The examiner scrutinizes the application to ensure that the invention is described through concrete technical concepts rather than vague or aspirational goals. For example, a claim that simply states “using AI to improve fuel efficiency” would likely be rejected for being too broad and lacking a specific technical character. Instead, the applicant must detail the specific parameters the AI monitors, the algorithms used to process that data, and how the output directly modifies the engine’s operation. This level of detail is necessary to distinguish a patentable invention from a mere idea. It forces the human inventor to demonstrate their mastery over the AI tool, showing that they have directed the machine toward a specific, repeatable, and useful result that represents a genuine advancement in the state of the art.
4. The CriteriFiltering Abstract Concepts and Theoretical Formulas
As the examination moves to the third step, the focus shifts toward ensuring that the proposal is not merely a vague or theoretical concept. Philippine law explicitly excludes “mathematical methods” and “abstract ideas” from patent protection, which poses a significant challenge for AI-related filings. Because AI is fundamentally built on complex mathematical structures and algorithms, the examiner must determine whether the invention provides a “technical solution to a technical problem” or if it is just a sophisticated way of performing mental acts. To pass this stage, the human inventor must prove that the AI’s involvement results in a tangible improvement to a technical process, such as increasing the speed of data transmission or enhancing the precision of a robotic arm. This requirement prevents the patenting of pure mathematics, ensuring that the building blocks of AI research remain available to the scientific community while protecting the specific, practical applications that drive economic growth.
The fourth step involves a fundamental review of the invention’s purpose and legal standing, specifically focusing on whether the use of AI respects public order and morality. This is where the mandatory scrutiny mentioned at the beginning of the article comes into play. In 2026, as AI is increasingly used in sensitive areas like cybersecurity and biotechnology, the IPOPHL must ensure that no patent is granted for an invention that could be used to facilitate illegal acts or violate the ethical standards of Philippine society. This step also involves checking if the invention falls into a recognized class of products or processes. The examiner looks for a “technical character” that transcends the mere execution of code. By verifying that the invention is not just a digital version of a well-known manual process, the office maintains the high standard required for an “inventive step,” which is the core requirement for any patent to be granted.
5. The Verification: Confirming Tangible Results and Solutions
The final step in the IPOPHL’s five-step analysis is the verification of tangible technical qualities or practical results. This is perhaps the most critical phase for AI-assisted inventions, as it requires the applicant to provide real-world evidence that the AI-driven solution actually works and produces a specific technical effect. Unlike traditional software, which can be evaluated based on its code, AI systems often function as “black boxes,” where the exact path from input to output is not always clear. To overcome this, the examiner looks for a physical embodiment or a specific technical application that solves a problem in a way that is not obvious to a person skilled in the art. Whether it is a more efficient way to manage power grids or a revolutionary method for early cancer detection, the invention must demonstrate a clear, functional benefit that goes beyond the theoretical capabilities of the AI model itself.
This rigorous verification process ensures that the Philippine patent system does not become a repository for “paper patents”—ideas that sound impressive but have no practical utility. In the competitive technological environment of 2026, this focus on tangible results is essential for maintaining the credibility of the patent office. It encourages inventors to move beyond the testing phase and develop mature technologies that can be deployed in the market. By requiring a physical or functional “embodiment,” the law anchors the digital world of AI to the physical world of industry. This approach not only protects the interests of legitimate innovators but also provides a clear roadmap for startups and researchers who are looking to commercialize their AI discoveries within the Philippines and abroad.
6. The Conflict: Addressing Infringement and Liability Challenges
One of the most complex issues facing the Philippine legal system today is the determination of liability when an AI-assisted invention is involved in a patent infringement case. Traditional legal tests, such as literal infringement and the doctrine of equivalents, were designed for scenarios where human actions could be easily traced and evaluated. Literal infringement occurs when a product or process includes every element of a patented claim, while the doctrine of equivalents applies when a substitute performs substantially the same function in substantially the same way to achieve the same result. When an AI system autonomously generates a solution that happens to mirror a patented invention, the question of “intent” becomes secondary to the technical reality of the infringement. However, identifying the responsible party remains a significant hurdle. In many cases, it is difficult to determine if the liability should fall on the developer who trained the AI, the owner who deployed it, or the user who provided the specific prompts that led to the infringing output.
The Philippine Supreme Court addressed some of these concerns in February 2026 through the Governance Framework on the Use of Human Centered Augmented Intelligence. This framework emphasized that human accountability must remain paramount in all AI applications. By imposing personal responsibility on the user for the output produced by an AI tool, the judiciary has signaled that individuals and corporations cannot hide behind the “autonomy” of their machines to escape legal consequences. Nevertheless, the burden of proof in these cases is exceptionally high. Discovering that a proprietary, closed-source AI system has infringed on a patent requires a level of technical forensic analysis that is often beyond the reach of standard litigation. As a result, legal experts suggest that the burden of proof may eventually need to shift, or new discovery rules may be required to allow for the inspection of AI training data and algorithmic structures without compromising trade secrets.
7. The Synergy: Data Privacy and Regulatory Compliance
In the current regulatory environment, the development of AI-assisted inventions is inextricably linked to the protection of personal data. The National Privacy Commission of the Philippines released updated guidelines in late 2024, emphasizing that any AI system processing personal data must adhere to the principles of transparency and accountability. For innovators, this means that the “training phase” of an AI-assisted invention is just as legally significant as the final product. If an inventor uses a dataset containing personal information—such as medical records or consumer behavior data—to train an AI that eventually produces a patentable invention, they must ensure that the initial data collection was lawful under the Data Privacy Act. A failure to comply with privacy regulations could not only lead to heavy fines but could also complicate the patent application process if the “prior art” or the methodology behind the invention is found to be based on illegally obtained information.
Moreover, the intersection of data privacy and patent law highlights the need for a holistic intellectual property strategy. While a patent protects the “inventive concept,” it does not necessarily protect the underlying data or the specific nuances of the AI model. In many cases, companies in the Philippines are adopting a hybrid approach, where they secure patents for the technical application of their AI while maintaining the proprietary datasets and optimized model weights as trade secrets. this strategy provides a broader safety net, ensuring that even if a patent is challenged or expires, the core competitive advantage remains protected. This multi-layered approach to innovation is becoming the standard for tech companies in 2026, as they navigate an increasingly complex landscape where data is the fuel and AI is the engine of progress.
8. The Framework: Judicial Oversight and Ethical Standards
The Philippine judiciary has taken a proactive role in shaping the legal context for AI innovation through the 2026 Governance Framework on the Use of Human Centered Augmented Intelligence. This framework is not just a set of internal rules for the courts; it serves as a guiding light for how AI should be treated across all legal domains, including intellectual property. The Supreme Court established that AI use must be human-centric, have a clear legal basis, and be proportional to the goal being achieved. This means that when a patent dispute involving AI reaches the court, the judges will look beyond the technicalities of the IP Code and consider the broader ethical implications of the technology. If an AI-assisted invention is found to be “unreasonable” or “disproportional” in its impact on society, it may face higher hurdles in enforcement or even be held to violate public order.
This judicial oversight ensures that the rapid pace of technological change does not outstrip the core values of the Philippine legal system. By emphasizing “augmented intelligence” rather than “artificial intelligence,” the court reinforces the idea that technology should enhance human capabilities rather than replace human judgment. This philosophy has a direct impact on how “inventive steps” are evaluated. If a machine does all the work without human guidance, the resulting output may be seen as a product of routine computation rather than an act of invention. The framework also underscores the importance of transparency, encouraging developers to create AI systems that are “explainable.” In a patent context, an explainable AI makes it easier for examiners and judges to understand the technical solution, thereby increasing the chances of a successful patent grant and providing a clearer path for defending that patent in court.
9. The Conclusion: A Foundation for Responsible Innovation
The legal community in the Philippines successfully recognized that the integration of artificial intelligence into the inventive process was not a threat to the patent system, but rather an opportunity to modernize it. By the middle of 2026, the implementation of the AI Examination Guidelines provided a necessary structure for distinguishing between human-led creativity and machine-generated outputs. The Intellectual Property Office of the Philippines and the National Privacy Commission worked in tandem to ensure that innovation did not come at the expense of ethical standards or data security. This coordinated effort allowed the Philippines to maintain its competitive edge in the global digital economy while providing clear, actionable pathways for inventors to protect their work.
Stakeholders within the technology sector realized that staying informed about these evolving regulations was the only way to safeguard their intellectual property assets. The shift toward a human-centered approach to AI ensured that the legal protections remained centered on the individuals who drove technological progress. As the legal landscape continued to mature, practitioners and innovators utilized these frameworks to build a robust ecosystem where AI served as a powerful instrument for national development. The proactive measures taken by the judiciary and the patent office ultimately fostered a culture of responsible innovation, ensuring that every AI-related patent filing was evaluated with the diligence required to uphold the integrity of the Philippine patent law system.
