In the rapidly evolving world of artificial intelligence, where innovation happens at breakneck speed and competition is fierce, how can growing AI companies protect their most valuable assets and secure their place in the market? Robert Plotkin’s new book “AI Armor: Securing the Future of Your AI Company with Strategic Intellectual Property” offers a compelling answer to this critical question.

As an experienced patent attorney specializing in software and AI, Plotkin brings decades of expertise to bear on the unique challenges faced by AI startups and growing companies. He argues convincingly that IP protection, when approached strategically, can be transformed from a mere defensive tool into a powerful driver of business growth and success.

The book’s central premise is that traditional approaches to IP are inadequate for the complex, fast-paced world of AI innovation. Instead, Plotkin introduces readers to his proprietary MIND process – a systematic framework for developing a comprehensive IP strategy tailored specifically to AI technologies and business goals.

The MIND Process: A Strategic Approach to AI IP

At the heart of “AI Armor” is the MIND process, which stands for Map, Identify, Navigate, and rive. This four-step approach provides a structured way for companies to:

1 . Map the unique aspects of their AI technology

The MIND process begins with a comprehensive mapping of your AI technology’s components, from training data to output processing. This systematic approach ensures that every innovative aspect of your AI system is identified and evaluated for its potential competitive advantage and IP protection.

2 . Identify strategic IP goals aligned with business objectives

Once your AI technology is mapped, the next step is to align your IP strategy with your broader business goals, such as fundraising, market dominance, or preparing for acquisition. This strategic alignment ensures that your IP efforts directly support and drive your company’s growth and success in the AI marketplace.

3 . Navigate the complex landscape of IP protection options

With a clear map of your AI innovations and strategic goals in place, the MIND process guides you through the intricate world of IP protection options. This step leverages specialized expertise to determine the most effective combination of patents, trade secrets, and other forms of IP protection for each component of your AI system.

4 . Drive successful business outcomes leveraging secured IP assets

The final stage of the MIND process focuses on actively leveraging your secured IP assets to achieve tangible business outcomes. This involves using your IP strategically to attract investment, generate revenue through licensing, increase company valuation for exits, and establish market leadership in the AI industry.

Mapping AI Innovations

The book places particular emphasis on the critical first step of mapping a company’s AI innovations. Plotkin introduces a comprehensive framework for breaking down AI systems into their core components, including:

  • Training data
    Training data is the foundation of any AI model, providing the raw material from which the model learns and derives its ability to make predictions or decisions. High-quality training data is crucial for developing accurate and reliable AI systems, whether it’s image datasets for computer vision, text corpora for natural language processing, or user interaction data for recommendation systems.
  • Training parameters
    Training parameters play a pivotal role in determining a model’s performance, affecting aspects like model complexity and the ability to generalize from training data to real-world scenarios. The process of fine-tuning training parameters often involves significant intellectual effort and can lead to substantial improvements in model accuracy and efficiency, making them prime candidates for IP protection.
  • Training module
    The training module is the engine room of machine learning systems, where the magic of learning from data unfolds to transform training data into a functional and intelligent trained model. Understanding and protecting the training module is critical for safeguarding AI innovations, as it often contains the heart of the novelty and competitive advantage conveyed by such systems.
  • Trained model
    The trained model represents the culmination of the training process, embodying the practical application and intelligence of the AI system. Its value lies in its functionality—whether interpreting complex data, making predictions, or automating decision-making processes—and contrary to common belief, trained models can be patentable when they offer novel functionalities or significant improvements over existing methods.
  • Model execution module
    The model execution module applies the trained model to live data to generate output, serving as the point where the rubber meets the road in AI systems. This module is integral to the functionality of AI systems and often involves significant effort to improve speed, output quality, or the ability to handle specific kinds of live data, making it a prime candidate for IP protection.
  • Model output
    Model output is the result of applying the trained model to live data and can take various forms, from specific predictions in data analytics to innovative designs for physical products. While it may seem like the least potentially patentable component, model output can represent inventions that may be patentable on their own, independently of how they were created.
  • Output processing
    The output processing module plays a pivotal role in transforming the initial model output into a form that is refined, usable, and suitable for commercial use. This module ensures that the raw output of the AI system meets necessary quality standards, ethical considerations, and legal requirements for real-world application, making it an important component to consider for IP protection.

From Defense to Offense: IP as a Business Driver

One of the book’s most valuable contributions is reframing how companies should think about IP protection. Rather than viewing patents and trade secrets as purely defensive measures, Plotkin demonstrates how a well-crafted IP strategy can actively drive business growth by:

  • Attracting Investment and Increasing Company Valuation
    Securing intellectual property (IP) demonstrates to investors that your company has a defensible competitive edge, making it a more attractive investment. A strong IP portfolio not only signals innovation but also increases company valuation by showcasing long-term growth potential and market differentiation.
  • Generating Licensing Revenue
    Licensing your patented AI technology to other companies creates an additional revenue stream without the need for direct market expansion. This approach allows you to monetize your innovations while retaining ownership, enabling you to fund further R&D or operational growth.
  • Facilitating Successful Exits Through Acquisition or IPO
    A robust IP portfolio enhances your company’s appeal to potential acquirers or public investors by showcasing proprietary assets that competitors cannot easily replicate. Patents and trade secrets can significantly boost acquisition offers or IPO valuations by demonstrating the scalability and exclusivity of your technology.
  • Attracting and Retaining Top Talent
    A well-protected innovation ecosystem signals to top talent that your company values creativity and is committed to safeguarding employee contributions. Additionally, employees are more likely to stay when they see their work being strategically leveraged for long-term success, creating a culture of trust and innovation.

He provides concrete examples of how AI companies have leveraged IP assets to achieve these outcomes, making a compelling case for investing in strategic IP protection early in a company’s lifecycle.

Debunking Common IP Myths

“AI Armor” also takes aim at several pervasive myths about IP protection in the AI industry. Plotkin systematically dismantles misconceptions such as:

  • AI and Software Can’t Be Patented
    The belief that AI and software can’t be patented is a widespread misconception that has deterred many innovators from seeking IP protection. In reality, AI and software have been patentable for decades, provided they meet the criteria of novelty, utility, and non-obviousness. Courts and patent offices worldwide have granted countless patents for software-based innovations, including AI algorithms and systems, as long as they solve a specific technical problem.
    Moreover, the scope of patentable subject matter in AI continues to evolve, with many jurisdictions recognizing the unique challenges posed by this technology. By working with experienced IP professionals, companies can craft strong patent applications that highlight the technical contributions of their AI innovations, ensuring robust protection in a competitive market.
  • First-Mover Advantage Is Sufficient Protection
    While being the first to market can provide an initial competitive edge, it is rarely enough to sustain long-term success in a fast-moving field like AI. Competitors with greater resources can quickly replicate or improve upon your innovations, eroding your market position. Without IP protection, there is little to prevent others from capitalizing on your hard work and investment.
    Patents and trade secrets offer a more durable form of protection by creating legal barriers that competitors cannot easily overcome. A strong IP portfolio not only safeguards your innovations but also enhances your ability to attract investors, negotiate partnerships, and maintain a dominant position in the market.
  • Patents Are Easily Circumvented in AI
    Some believe that patents are ineffective in the AI space because competitors can easily design around them. While it’s true that poorly drafted patents may leave loopholes, well-crafted patents can provide broad and enforceable protection. By focusing on the unique technical aspects of an innovation and anticipating potential workarounds, companies can secure patents that are difficult to circumvent.
    Additionally, a comprehensive IP strategy often involves layering different forms of protection—such as patents for core technologies and trade secrets for proprietary data or methods—to create a robust defensive moat. This multi-faceted approach makes it significantly harder for competitors to replicate your innovations without infringing on your rights.
  • Patents Are Only for Big Companies With Deep Pockets
    The notion that patents are only accessible or valuable to large corporations is outdated and misleading. While it’s true that obtaining and enforcing patents requires an investment, there are cost-effective strategies available for startups and small businesses. For example, focusing on key innovations and filing strategically can maximize the return on investment in IP protection.
    Moreover, patents can be a powerful tool for smaller companies to level the playing field against larger competitors. They provide leverage in negotiations, attract investors by demonstrating defensibility, and even create opportunities for licensing revenue or strategic partnerships. Many successful startups have used their patent portfolios as a cornerstone of their growth strategy.
  • The Patent Process Is Too Slow for Fast-Moving AI Innovation
    Critics often argue that the traditional patent process is too slow to keep pace with the rapid advancements in AI technology. While it’s true that obtaining a patent can take time, expedited examination programs are available in many jurisdictions to accelerate the process for high-priority applications. In some cases, patents can be granted in under a year through these fast-track options.
    Furthermore, filing a patent application early establishes a priority date, which provides immediate protection against subsequent filings by competitors. Even as technology evolves, having foundational patents in place ensures that your core innovations remain protected while you continue to innovate and adapt to market changes.

By addressing these myths head-on, the book empowers readers to make more informed decisions about IP strategy and avoid common pitfalls.

Practical Guidance for AI Leaders

While “AI Armor” delves into complex legal and technical concepts, it remains accessible and practical throughout. Plotkin excels at translating his expertise into actionable advice for founders, executives, and investors in AI companies.

The book is peppered with real-world case studies and examples that bring key concepts to life. Plotkin draws on his extensive experience working with AI clients to illustrate both successful IP strategies and cautionary tales of companies that failed to adequately protect their innovations.

Particularly valuable are the detailed checklists and frameworks provided for each stage of the MIND process. These tools give readers a concrete starting point for developing their own IP strategies, even if they lack deep legal or technical expertise.

The Urgency of Strategic IP Protection

A recurring theme throughout “AI Armor” is the time-sensitive nature of IP protection in the AI industry. Plotkin makes a compelling case that the window of opportunity for securing valuable IP rights is limited, especially given the rapid pace of innovation and increasing competition from both startups and tech giants.

He emphasizes that waiting to implement an IP strategy until after achieving significant growth or market traction can be a costly mistake. By that point, it may be too late to protect key innovations due to legal deadlines or competitors beating you to the patent office.

This sense of urgency is balanced with practical advice on how to develop a cost-effective IP strategy, even for early-stage companies with limited resources. Plotkin outlines approaches for prioritizing protection efforts and maximizing ROI on IP investments.

Beyond Patents: A Holistic View of IP Protection

While patents feature prominently in the book, “AI Armor” takes a holistic view of IP protection. Plotkin emphasizes the importance of considering multiple forms of IP rights, including:

  • Patents
  • Trade secrets
  • Copyrights
  • Trademarks

He provides guidance on when each type of protection is most appropriate and how different IP rights can be combined to create a comprehensive defensive moat around a company’s AI innovations.

The book also dives into strategies for protecting non-patentable aspects of AI systems, such as training data and business methods. This nuanced approach recognizes the multifaceted nature of AI technologies and the need for tailored protection strategies.

Preparing for the Future of AI

“AI Armor” doesn’t just focus on the current state of AI and IP law. Plotkin also looks ahead to emerging trends and potential future developments that could impact IP strategy for AI companies. He discusses topics such as:

  • The Potential Impact of AI-Generated Inventions on Patent Law: AI-generated inventions challenge traditional patent frameworks by raising questions about inventorship, ownership, and eligibility. As AI systems increasingly contribute to innovation, patent laws may need to evolve to address whether AI can be recognized as an inventor or how human-AI collaboration should be treated in the patenting process.
  • Evolving Regulatory Landscapes for AI Technologies: The regulatory environment for AI is rapidly changing, with governments worldwide introducing new frameworks to address ethical, safety, and transparency concerns. These evolving regulations can impact how companies protect their intellectual property, requiring them to stay agile and compliant while safeguarding their innovations.
  • Ethical Considerations in AI Development and Their IP Implications: Ethical considerations in AI development, such as bias mitigation and data privacy, can influence the scope and nature of intellectual property protection. Companies must balance their IP strategies with ethical responsibilities to ensure their innovations align with societal values and regulatory expectations.

This forward-looking perspective helps readers future-proof their IP strategies and anticipate potential challenges on the horizon.

A Must-Read for AI Leaders

“AI Armor” is an essential read for anyone involved in leading or investing in AI companies. It fills a critical gap in the literature by providing a comprehensive, practical guide to IP strategy specifically tailored to the unique challenges of the AI industry.

Plotkin’s deep expertise shines through on every page, but the book remains accessible and engaging throughout. He strikes an effective balance between high-level strategic insights and detailed tactical advice, making the book valuable for both C-suite executives and technical leaders.

While the legal and technical concepts covered are complex, Plotkin’s clear writing style and liberal use of real-world examples keep the material grounded and applicable. Readers will come away with a solid foundation for developing their own IP strategies, as well as a keen appreciation for the strategic importance of IP in driving business success.

Conclusion: Securing Your AI Company’s Future

In an industry where innovation is the lifeblood of success, protecting that innovation is paramount. “AI Armor” makes a compelling case that strategic IP protection is not just a legal necessity, but a powerful tool for driving growth, attracting investment, and securing long-term competitive advantage.

By providing a systematic framework for developing AI-specific IP strategies, along with practical guidance for implementation, Plotkin has created an invaluable resource for the AI industry. For any company looking to thrive in the competitive world of AI innovation, “AI Armor” offers a roadmap for turning intellectual property from a cost center into a strategic asset.

In a field where the difference between success and failure often comes down to who can innovate fastest and protect their innovations most effectively, the strategies outlined in “AI Armor” may well prove to be the decisive factor in determining which AI companies rise to dominance and which fall by the wayside. For leaders serious about securing the future of their AI companies, this book is required reading.

Interview with the author Robert Plotkin

1 . Why did you write the book?

I wrote the book to help AI companies understand the significant benefits they could obtain from pursuing strategic intellectual property protection for their innovations, especially using patents and trade secrets. In my nearly 30 years of working with innovative tech companies, I have found that many founders, executives, and investors act based on common myths about intellectual property (IP), and as a result they fail to take advantage of the ways in which IP can help them to attract investment, protect them against competitors, attract scarce talent, and boost their value for an acquisition, IPO, or other exit. I wanted to share the strategies that I have used successfully with my startup clients over many years so that other companies could benefit from them, especially early-stage AI companies who are developing significant innovative AI technologies that need IP protection to support the growth of the company and to protect against much larger competitors.

2 . Who is this book particularly suitable for – what do you think is the greatest added value of the book for the target group?

I wrote the book specifically for AI startup company founders, executives, investors, and advisors, although it is still valuable for anyone involved in a growing innovative technology company.

The book is unique in that it provides a clear and systematic process for evaluating the different forms of IP protection that are available for AI innovations, including a decision process of choosing the best forms of IP to use for achieving specific business goals. Any person at a tech company can understand the step-by-step process described in the book and use it to guide strategic decisions about IP. This includes deciding among patent, trade secret, and copyright protection, based on the particular features of the company’s technology and the company’s business goals.

One significant benefit of the book is that it describes how to use a company’s business goals to drive IP strategy, so that the IP strategy supports and promotes those business goals. Too often, IP attorneys apply a “cookie cutter” approach to each client’s IP, which often results in IP that does not align with the client’s specific needs. For example, many companies obtain patents that are too narrow or that don’t cover the key features of the company’s core innovations. The book addresses this problem by explaining how to develop IP that closely aligns with the company’s business goals and innovations, so that the IP can have maximum value for driving those goals.

3 . Why do you think so many founders, entrepreneurs and business decision-makers still find it so difficult to understand the basics of IP?

Too many IP professionals describe IP using highly-legalistic terminology and focus on the detailed mechanics of IP, which can make IP difficult for anyone other than an IP lawyer to understand.

As an analogy, when most people buy an automobile, they want to know how it will feel to drive it, whether it is safe, and whether they and their family will be comfortable in it. They want to know whether it will do the job of getting them to work or the grocery store easily, safely, and enjoyably. If the salesperson only describes the engine, drive train, and exhaust system of the car to the customer, it will be very difficult for the customer to understand whether to buy the car. Yet this is what IP professionals do all the time when they explain IP to entrepreneurs. So it is no surprise that entrepreneurs find it difficult to understand the basics of IP.

Startup decision-makers are smart people and they can easily understand IP if it is explained to them in business terms. This is what I strived to do in the book.

4 . In your opinion, what kind of game-changing potential does AI have for the commercial use of IP?

Let me answer this from two directions:

(1) IP protection for AI has the potential for broad commercial value generation

Because AI is useful for so many different applications and in so many fields, innovators who obtain foundational patents now on core AI algorithms and broadly-applicable AI innovations have the potential to license those patents extensively to generate significant revenue, or to use those patents to support businesses that can capture significant market share. Whenever a new field of technology arises or there is a rapid phase of innovation in an existing technology, opportunities arise to obtain “pioneer patents” that can become the foundations of entire industries for years or decades into the future. I believe we are in that period of time now. This is exciting and also nervewracking, because the window of opportunity to obtain such patents often closes after some number of years, once the key innovations have been developed and patent protection is no longer available for them. At that time it is still possible to obtain narrower patents that are still valuable, but not as broadly applicable.

(2) AI has game-changing potential for use in commercialization of IP

IP is embodied in the form of legal rights, which typically are defined by textual legal documents which describe the scope of rights. Today’s AI systems, especially those based on large language models (LLMs), excel at understanding, interpreting, and drawing conclusions based on text. AI, therefore, provides an increasingly powerful ability to help IP owners:

  • understand what IP they own;
  • find infringers of their IP;
  • develop evidence of infringement;
  • uncover commercialization opportunities for IP; and
  • even evaluate potential new IP.

Before the Internet and AI, think about how difficult it was for a patent owner to determine whether anyone was infringing the patent. If the patent was for a mousetrap, it could be very difficult to find someone selling an infringing mousetrap in a remote part of the country. However, now:

  • The Internet makes information about potentially infringing products widely and instantly available.
  • AI makes it possible to search for, interpret, and evaluate such product information for infringement quickly and inexpensively.

This is just one example of how AI has game-changing potential for commercialization of IP. The combination of increasingly-available commercial information online and the ability of AI to automatically, quickly, and inexpensively analyze information  about IP owners’ own technologies and the technologies of other entities opens up vast new opportunities for leveraging AI to commercialize IP.

Thank you, Robert, for the interview!