Back to Trends
Reading Time:
15min

Protect Your IP from AI Danger

March 11, 2024
Artificial Intelligence

In a hectic tech transfer office, with stacks of invention disclosures, trying to give each one the attention it deserves can feel overwhelming. Every day, there's more to look at – a potential world changing breakthrough. And there you are, swamped, wondering how you're going to get through it all.

Then, in a quiet moment, you catch yourself scrolling through your phone and boom – you see all these raves about generative AI being a game-changer. Could this be the solution for your overwhelm? The thought of using generative AI to sift through the initial ideas, spot the gems, and maybe even whip up some great marketing summaries starts to feel less like a daydream and more like a seriously smart move.

But before diving headfirst into the generative AI pool, there's a big problem you must pay attention to: the patent filing and publication status of your inventions. It's like navigating a minefield blindfolded if you're not careful. Just because AI can, doesn't mean AI should—not in all circumstances, at least. This isn't just about keeping your valuable inventions safe from copycats; it's about making sure you're not accidentally giving away the secret sauce.

So, as much as you're itching to let AI make your life easier, you need to review each invention’s patent filing and publication status first.

To make it easier, I’ve created a quadrant diagram below. The vertical axis relates to the publication status, with “published” on top and “secret” at the bottom. The horizontal axis relates to the patent filing status, with “not filed” at the left and “patent pending” at the right. Only inventions that are in the top right quadrant – both published and patent pending – are safe to put into generative AI models. Here’s why.

Understanding the Quadrants of IP Protection

IP Protection Quadrants - described below

Understanding the quadrants of IP protection is crucial for navigating the complexities of how your invention is safeguarded, particularly in the context of prior art and the nuances of international patent law. Once your invention has reached the patent pending status, it benefits from the protection of the filing date, which is a significant milestone in securing your intellectual property rights. However, the treatment of a patent application that is filed but not yet published varies across different jurisdictions – especially as a blocker for future patent applications.

Normally this wouldn't matter, as your unpublished invention should remain secret and unavailable to other innovators. But when you upload your invention to generative AI, you're exposing it to potentially being used as training material. This means that elements of your invention could be remixed into another person's generative AI produced content.

This situation raises complex questions about ownership and whether an unpublished, though filed, patent application provides sufficient protection in the face of generative AI’s capacity to remix and repurpose IP. It underscores the importance of carefully considering the status of your IP and the potential implications of engaging with generative AI technologies on the global stage.

For more information, you can read our detailed review of publishing fast vs filing fast in the age of generative AI.

Next up: what this means for you practically speaking.

Upper Right Corner: Patent Pending and Published - The Ideal Scenario

For your TTO, having an invention that is both patent pending and published represents the gold standard of IP protection. This status not only fortifies the invention against potential infringement but also establishes a strong position for licensing negotiations. The dual advantage of a patent filing date and the broad disclosure through publication serves as a robust deterrent against competing inventions, ensuring that your TTO retains a competitive edge.

In the age of generative AI, it also means that you can protect your IP from the effects of remixing. Getting a granted patent means that you can prevent others from making, using, selling, etc your idea. Publishing fast means that others will have a harder time getting a patent in a similar or adjacent area – plus you’re contributing to the world’s knowledge and progress. A win-win!

Lower Right Corner: Patent Pending but Not Published - A Cautionary Position

Although securing a patent pending status is a significant step toward protecting an invention, the absence of publication introduces vulnerabilities. Tech transfer organizations should be cautious, as the lack of public disclosure may weaken its position as prior art. If you upload your idea to generative AI, your protection against remixing is potentially more limited. In this scenario, the recommendation is to publish before using generative AI, to maximize protection and market potential.

Upper Left Corner: Published but Not Patent Pending - Public Domain Risks

When an invention is published but not patent pending, it enters the public domain, making it accessible for AI training purposes. While this might seem advantageous for visibility, it poses a significant risk: the invention's unique elements could be reimagined into new inventions that others might patent. For tech transfer organizations, this means any subsequent developments or applications derived from the original invention could be legally out of reach. To mitigate this risk, considering patent filing before publishing – or file within one year of publication (US only).

Lower Left Corner: Neither Published Nor Patent Pending - High Vulnerability

Inventions that are neither published nor patent pending are at the highest risk when exposed to consumer generative AI models. For tech transfer organizations, uploading such inventions to AI platforms is fraught with danger, as it exposes the IP to potential use, remixing, and even patenting by third parties without any form of protection. This quadrant represents a scenario to be avoided, underscoring the importance of securing either publication or patent pending status before engaging with generative AI technologies.

Mastering the AI Wave: Smart IP Strategies for Tech Transfer Success

For tech transfer teams, leaning into generative AI to find the best licensing leads and whip up some sharp market summaries has a lot of upside. It’s all about timing and making sure your IP game is strong. Diving into AI can totally change the game for pushing research breakthroughs out into the world. Still, you need to be careful. Balancing the need to protect your IP while getting the most out of AI's nifty tricks is the name of the game.

We'll like to get your thoughts on using AI tools for tech transfer - so please comment below.

Plus, we're putting together a course on The Future of Tech Transfer Marketing: Leveraging AI for Licensing Deals. You can check out the course here. Please give us your feedback - and of course, sign up!

Also don't miss:
Reading Time:
15min

AI As Corporate Ozempic - Boosting Efficiency or Eroding Skills?

AI can streamline operations but risks "deskilling" human abilities. Use AI to enhance, not replace, critical thinking and creativity.

August 23, 2024

Learn More...

Artificial Intelligence
Reading Time:
15min

Natural Language Processing (NLP) and speech tasks

Natural Language Processing has speech tasks that enable users to speak to any device. In this post, we explore how that works (with examples!)

June 26, 2023

Learn More...

Artificial Intelligence
Reading Time:
15min

Why It Isn’t Always About AI

Artificial Intelligence is not a buzzword, it’s a tool and, although it’s very useful, it’s not necessary to use it every time when you want to solve a problem, therefore it isn't always about AI.

June 26, 2023

Learn More...

Artificial Intelligence