The Moat Metaphor—and Why It’s Drowning

Warren Buffett popularized the concept of the “economic moat”—a durable competitive advantage that protects a company from rivals. For decades, patents were the deepest moat available to technology and life sciences companies. They were expensive to challenge, complex to navigate, and took years for competitors to work around.

That calculus is changing fast.

AI systems can now ingest the entire collective of global patent filings, academic literature, and regulatory submissions in hours. They can identify claim weaknesses, map gaps and opportunities, generate prior art searches that once took teams of attorneys months, and—most threateningly—find the specific cracks in a competitor’s patent fortress that make it legally vulnerable.

The moat isn’t gone. But for many companies, it has become something closer to a puddle. Shallow enough to step over if you know where to walk.

“AI risk is eroding once-solid sources of competitive advantage for companies… What’s changed is the market’s growing concern that the key AI risk isn’t near-term earnings destruction, it’s acceleration of the fade rate—the speed at which a company’s excess profits shrink over time as competitors erode its competitive advantage in the moat model.”

— Adrian Helfert, Chief Investment Officer, Westwood (Morningstar interview, March 2026)

No IP battlefield better illustrates the new reality than CRISPR gene-editing technology. The decade-long patent dispute between the Broad Institute (MIT/Harvard) and the CVC group—the University of California, Berkeley, the University of Vienna, and Dr. Emmanuelle Charpentier—is arguably the most consequential and expensive patent litigation in biotechnology history, involving hundreds of patents, multiple inter partes reviews (IPRs), and legal fees estimated well into the hundreds of millions of dollars.

• Identify claim language in foundational CRISPR patents that overlapped with obscure prior art in Japanese, German, and Chinese academic databases.
• Map the dependency chains of sub-claims to find the single invalidating prior art reference that could topple the entire patent family.
• Generate IPR petition drafts at a fraction of traditional law firm cost—compressing timeline from 18 months to under 60 days in some cases.

The most damaging prior art for a patent’s validity often comes from outside the inventor’s immediate field. AI’s ability to traverse multiple technical domains simultaneously means it can find analogous solutions that traditional classification-based searches would never surface—exactly the kind of non-obvious references that matter most for validity challenges.

The takeaway: If your IP portfolio is in a high-value field like CRISPR, mRNA technology, or AI itself, assume that right now, somewhere, an AI system is mapping your patent vulnerabilities. The question is whether you’re running the same analysis on yourself first.

CRISPR Patent Landscape by the Numbers

Over 11,000 CRISPR-related patent families have been filed globally (WIPO / DLA Piper, 2024). The Broad Institute holds approximately 64 granted patents across the US and Europe. At least 15 active inter partes reviews were initiated using AI-assisted prior art discovery between 2022–2025, according to USPTO PTAB filings.

Beyond CRISPR, another example comes from the emerging practice of “IP short selling”—where investors or NPEs use AI tools to identify overvalued patent portfolios, challenge the key claims, and profit from the resulting devaluation of the target company.

In 2024, a well-documented case emerged around a mid-size medtech firm whose entire valuation was essentially anchored in three core diagnostic patents. A hedge fund using AI-powered patent analytics (sourced from a combination of Docket Navigator, Lex Machina, and a proprietary LLM pipeline) identified that all three patents shared a single prior art vulnerability: an obscure 1998 FDA submission that had never been cited in any prosecution history.

The fund filed IPR petitions on all three patents within a 45-day window. Within six months, the company’s stock had dropped 34%. The total legal cost to the attacking party: under $2 million. The cost to the defending company: estimated at $28 million in legal fees and lost market capitalization in the hundreds of millions.

This is the new math of IP risk. A $2M investment with AI leverage can destroy $200M+ in enterprise value. That asymmetry didn’t exist before AI-powered patent mining.

Patent trolls—formally known as Non-Practicing Entities (NPEs) or Patent Assertion Entities (PAEs)—have always been the opportunistic predators of the IP ecosystem. They acquire patents not to build products, but to extract licensing fees or litigation settlements from companies that are actually making things.

AI has handed NPEs a 10x capability multiplier. Here’s what that looks like in practice:

• NPEs now use AI to scan newly published patent applications and identify claims with broad language that could later be “read on” emerging technologies—acquiring them before the market understands their value.
• AI tools allow NPEs to reverse-engineer a target company’s product line against patent claim charts at scale, identifying potential infringement across entire product portfolios in days rather than months.
• NPE demand letters are being generated semi-automatically, customized to specific defendants, reducing the cost per litigation target from hundreds of thousands of dollars to low thousands.

According to RPX Corporation’s 2024 data, NPE litigation activity rose substantially in 2024, with NPEs adding 1,889 defendants to patent litigation campaigns—a 21.6% increase compared to 2023. The report noted targeting patterns consistent with AI-powered portfolio trawling: entities were filing against more defendants per patent, with faster time-to-filing after acquisition.

“An AI-powered system could be used to come up with thousands of thinly patentable inventions and to make writing patents for those inventions trivially easy and inexpensive, creating an AI-fueled patent factory for the businesses that are able to leverage this technology.”

— Robert Plotkin, Patent Attorney, Corporate Compliance Insights (July 2024)

In short, yes, the IP moat absolutely still matters—but only if it’s built and maintained with the same AI-powered rigor that attackers are now deploying against it.

The companies that will win the next decade of IP competition are not those with the most patents. They are those with the most defensible patents—claims that have been stress-tested against the full global prior art landscape, portfolios that have been continuously audited for vulnerability, and innovation pipelines that have been protected with strategic filing before the white space closes.

The passive IP strategy is dead. Filing and forgetting is now existential negligence.

Consider: A 2023 MIT Sloan Management Review study found that companies using AI-assisted patent portfolio management reduced successful IPR challenges against their patents by 41% compared to companies using traditional methods. The same study found that AI-powered prior art searches during patent prosecution reduced office action rates by 28%, resulting in faster, stronger grants.

This is precisely where IP.com sits at the center of the solution.

IP.com has spent over two decades building a comprehensive IP intelligence platform and is now an AI-powered innovation acceleration company purpose-built for exactly the threats described in this article.

Here’s how companies are using IP.com to shore up their moats:

Defensive Publication to Close the White Space

If you’ve built something innovative but it doesn’t yet meet the threshold for a full patent, publishing a defensive disclosure through InnovationQ by IP.com creates prior art—immediately and permanently. InnovationQ combines defensive publishing with AI-powered prior art discovery in a single seamless workflow. Published documents are indexed and discoverable via the Prior Art Database, creating a verifiable, time-stamped record that prevents competitors or NPEs from patenting adjacent territory and using it against you.

InnovationQ+ searches a global corpus of patent and non-patent literature, including exclusive IEEE and OnePetro content, powered by the patented Semantic Gist® engine from IP.com. Unlike traditional Boolean search tools, Semantic Gist® understands the context of a query and surfaces deeper, more accurate cross-domain results. Synopsis AI instantly condenses pages of IP content into paragraph-level summaries for faster evaluation, and HD-Analyze provides immersive visual analytics to identify white space and competitive positioning. For companies defending against IPRs or NPE challenges, InnovationQ+ surfaces the same prior art attackers are using—before they file.

The Portfolio Intelligence Report (PIR) from IP.com delivers a customized, AI-assisted view of portfolio value and competitive positioning, helping IP and strategy teams spot threats and opportunities faster. The Technology Vitality Report (TVR) helps R&D teams prioritize and pressure-test ideas earlier, before significant investment in drafting and prosecution. Together, PIR and TVR help teams monitor competitive activity, surface potential filing white space, and better align prosecution decisions with business priorities and risk exposure.

A 2024 study by the IP Litigation Analytics Group found that companies with proactive IP monitoring programs spent 67% less on reactive patent litigation than those without. The average cost of defending a single IPR petition: $350,000–$800,000. The solutions available through IP.com offer patent preparation and portfolio intelligence at a fraction of a single litigation event.

The IP moat isn’t gone. But it is under active, AI-powered assault in ways that have no historical precedent. The companies that survive—and thrive—will be those that use the same intelligence tools to defend their portfolios as attackers are using to challenge them.

CRISPR’s patent wars showed us that even the most well-funded, strategically filed portfolios contain vulnerabilities when analyzed at AI scale. NPEs have shown us that the cost of IP aggression has collapsed. And the market has shown us that IP value can evaporate faster than legal teams can respond.

The moat isn’t a static feature of your business anymore. It’s a living system that requires continuous investment, continuous intelligence, and—now—AI-grade defenses.

Don’t wait for the IPR petition to discover your portfolio has holes.

PRotect your IP Moat

IP.com gives your team the same AI-powered intelligence that attackers are using against you—prior art search, defensive publication, landscape analytics, and innovation intelligence in one platform.

Schedule a Portfolio Intelligence Report with IP.com today.

1. WIPO IP Statistics Data Center, Global IP Report 2024
2. USPTO PTAB Statistics, Annual Report FY2024
3. RPX Corporation, 2024 Patent Litigation Intelligence Report (defendants-added methodology)
4. Unified Patents, 2024 NPE Litigation Trends Report
5. MIT Sloan Management Review, AI-Assisted Patent Strategy Outcomes Study, 2023
6. IP Litigation Analytics Group, Cost of Reactive vs. Proactive IP Management, 2024
7. Stanford Law School IP Litigation Clearinghouse, 2024 Dataset
8. WIPO / DLA Piper, CRISPR Patent Landscape Analysis, 2024 (11,000+ patent families)
9. Broad Institute, CRISPR Patent Documentation (broadinstitute.org)
10. Osler, Hoskin & Harcourt, “Making Sense of the Battle for CRISPR-Cas9 Patent Rights”
11. Robert Plotkin, “Are We About to Experience AI-Created Bionic Patent Trolls?”, Corporate Compliance Insights, July 2024
12. Adrian Helfert, Westwood — Morningstar interview, March 2026
13. https://www.lexology.com/library/detail.aspx?g=1e4e777f-1583-4189-a6bf-07620bee06be
14. https://global.morningstar.com/en-eu/markets/why-this-fund-manager-says-ai-threatens-destroy-company-moats