The Code War: AI Puts Standards and Copyright on a Collision Course

NEW YORK, NY – June 29, 2026 – On the surface, the American National Standards Institute (ANSI) is hosting a routine legal forum in Denver next month. A press release announces the gathering of executives, lawyers, and policy leaders to discuss artificial intelligence. But reading the underlying signals reveals this is no ordinary conference. The event, titled "Standards at a Crossroads," is a thinly veiled war council. The industry that creates the foundational rules for everything from aerospace engineering to medical devices is confronting an existential threat, and it is deeply, dangerously divided on how to fight back.

The adversary is the very technology poised to define the next century: artificial intelligence. For the organizations that develop, curate, and sell access to technical standards, AI is a double-edged sword of immense proportion. It promises to unlock unprecedented value from their dense, complex content, yet it also threatens to consume and commodify their intellectual property, potentially gutting the business model that has sustained them for decades. As ANSI President and CEO Laurie E. Locascio noted, "AI is rapidly changing the commercial realities of standards." That single sentence belies a roiling sea of confidence and concern, ambition and anxiety, that will converge at the Grand Hyatt Denver.

A Fractured Defense: The SDO Dilemma

The core of the conflict lies in the divergent strategies of the Standards Developing Organizations (SDOs) themselves. An investigation into their current policies reveals not a united front, but a fractured landscape of conflicting philosophies. This is the first signal of weakness the industry is broadcasting.

On one extreme is HL7 International, the steward of healthcare data standards. It has thrown its gates wide open, engineering its flagship FHIR standard to be explicitly "AI-ready" and publishing it under a Creative Commons Zero (CC0) public domain license. This is a radical bet on open adoption, viewing AI developers as valuable consumers to be courted, not a threat to be contained. The intent is clear: sacrifice direct control to achieve maximum influence and ubiquity in the burgeoning healthcare AI sector.

In stark opposition stand the old guards like ASTM International and SAE International, which oversee standards for materials, manufacturing, and mobility. Their terms of use are unequivocal, explicitly prohibiting the use of their standards to train AI models without permission. They are building a fortress, partnering with entities like the Copyright Clearance Center to manage a tightly controlled flow of information. Their stance signals a deep-seated belief in the intrinsic, licensable value of their IP and a profound distrust of AI platforms that scrape data indiscriminately.

Occupying the middle ground is a third faction, exemplified by the International Code Council (ICC). Rather than opening the gates or building a fortress, the ICC has created its own “walled garden.” It developed a proprietary generative AI tool, the AI Navigator, trained exclusively on its own copyrighted building codes. This strategy attempts to have it both ways: leveraging AI's power to enhance user experience while keeping the underlying IP—and the value generated from it—securely in-house. This move signals a strategic intent to become a tech provider, not just a content publisher, controlling both the code and the means of its interpretation.

These three distinct strategies—open adoption, strict prohibition, and proprietary integration—cannot coexist indefinitely. The ANSI forum's first panel, aptly named "What SDOs Are Doing," will put leaders from these rival camps on the same stage. The ensuing discussion will be less a panel and more a public negotiation over the very soul of the standards industry.

The Commercial Battlefield: Innovation vs. Monetization

Fueling this internal SDO conflict is immense external pressure from the market. The second panel at the ANSI forum brings the builders to the table—representatives from Amazon Web Services, Accuris, and other system integrators who are actively coupling digitized standards with AI platforms. Their goal is to create new products and commercial opportunities, turning static codebooks into dynamic, intelligent tools.

The demand is undeniable. Architects want AI that can instantly validate designs against thousands of pages of building codes. Engineers need systems that can cross-reference material specifications and safety standards in real time. These use cases are not theoretical; the ICC’s AI Navigator has already processed over 140,000 inquiries, demonstrating a voracious appetite for AI-powered standards access. This is the commercial opportunity that SDOs are desperate to capture.

But here lies the peril. To enable these innovations, SDOs must digitize their content and make it machine-readable. In doing so, they risk losing control. Once a standard is ingested by a large language model, its essence can be regurgitated, summarized, and applied without the user ever paying a licensing fee or even knowing the original source. This devalues the SDO's core asset. The very act of enabling innovation could trigger their own obsolescence. This is the tightrope SDOs must walk: how to license their content for AI use in a way that fuels new growth without cannibalizing their existing revenue.

The Search for New Rules in a Lawless Land

Underpinning the entire debate is a profound legal vacuum. The current intellectual property framework is ill-equipped for the age of generative AI. The U.S. Copyright Office maintains that only works with human authorship can be protected, leaving the output of sophisticated AI in a legal gray area. Meanwhile, landmark court cases are currently debating the fundamental question of whether training an AI model on copyrighted data constitutes fair use.

The ANSI forum's third panel, "How SDOs Can Protect Their IP," is a direct acknowledgment of this legal chaos. The inclusion of experts from the U.S. Patent and Trademark Office, the Copyright Clearance Center, and top law firms signals an urgent search for practical tools in the absence of clear legal precedent. The discussion is set to focus on contract frameworks, protective provisions, and liability allocation—the building blocks of a new legal order that SDOs must construct for themselves.

They are not just discussing theory; they are attempting to draft the treaties that will govern the frontier between legacy knowledge industries and the new world of generative AI. The outcomes of these discussions will provide a playbook for any industry whose value is rooted in proprietary, text-based information—a category that includes legal services, scientific publishing, and journalism.

By convening this summit, ANSI is playing a high-stakes game. It is publicly exposing its community's deep divisions and vulnerabilities. But in doing so, it is also asserting its role as the indispensable mediator. It has brought the factions to the table: the guardians of the old order, the builders of the new, and the lawyers trying to write the rules of engagement. The forum in Denver is more than a meeting; it is the opening negotiation for the future of how knowledge is created, controlled, and monetized in the age of artificial intelligence.