Sidney Wright, J.D. Class of 2027
Patents have long been the cornerstone of intellectual property protection, rooted in decades of precedent and a long-standing belief that ownership drives innovation. Patents have offered inventors exclusivity in exchange for transparency, shaping how technology has been created and shared. But as artificial intelligence (“AI”) becomes more embedded in everyday life, companies crave protection that meets the pace and nature of the technology they build.
Patent law requires that an invention meet specific thresholds of eligibility, novelty, usefulness, and non-obviousness. However, as cases like Recentive Analytics v. Fox demonstrate, courts increasingly hold that simply applying AI or machine-learning techniques to familiar processes is not enough to make an invention patentable. To be patentable, a claim must demonstrate a concrete improvement in how the AI itself functions, not just its application to a new field. For many companies, that threshold is hard to meet, and even when it is, the lengthy patent process often outlasts the commercial life cycle of the technology.
The movement toward trade secrets offers new forms of flexibility and control but also introduces its own set of limits and vulnerabilities. Observing this shift raises questions not only about how intellectual property law adapts to technological change, but also about what that adaptation might signal for the future.
Why Companies Are Turning Toward Trade Secrets
Since the passage of the Defend Trade Secrets Act (DTSA) in 2016, trade secret filings have risen steadily as patent litigation has declined. Within a year of its passage, trade secret cases increased by 25%, and last year more than 1,200 were filed in U.S. courts.
For many companies, especially those operating in fast-moving industries like AI, trade secret protection offers a combination of practicality and adaptability that traditional patent systems struggle to match. From a procedural standpoint, trade secrets provide immediate and cost-effective protection. There is no application process, waiting period, or public disclosure requirement, and the protection can last indefinitely as long as confidentiality is maintained. In contrast, the patent process can take years, often outlasting the commercial life of the technology it seeks to protect. For businesses that rely on speed, flexibility, and innovation, trade secret protection aligns more naturally with how modern research and development operates.
The strategic fit is especially apparent in the AI sector, where much of what gives a company its competitive edge, such as training data, model architecture, tuning methods, or even the knowledge of what approaches failed, is not patentable under current legal standards. Trade secrets, in contrast, can safeguard the inner workings of algorithms or data sets without requiring public disclosure. They also avoid the growing risk of invalidation that many patents face under §§ 101 and 112 of the Patent Act, which have narrowed the scope of patentable subject matter.
Beyond these legal and technical considerations, trade secrets offer companies greater control over how their intellectual property is managed. Rather than relying on public filings and external review, protection depends on internal safeguards such as confidentiality agreements, access restrictions, and data security measures, which companies can tailor to their specific needs.
Limits of Relying on Trade Secrets
While trade secret protection offers flexibility and speed, it also carries significant vulnerabilities. Unlike patents, trade secrets provide no recourse once information becomes public, whether through a leak, a data breach, or even unintentional disclosure. The protection ends the moment secrecy is lost, and there is no formal registry to prove prior ownership. In industries like AI, where products can be reverse-engineered or independently developed, this risk is particularly high. Maintaining secrecy also requires companies to meet a demanding and evolving “reasonable measures” standard, which may include rigorous cybersecurity, employee training, and data-access controls. Litigation over misappropriation can be costly and complex, and the discovery process itself may expose the very information a company sought to protect.
What This Shift Might Mean for the Industry
As more companies rely on trade secret protection, legal teams are spending greater attention on confidentiality controls, employee agreements, and cybersecurity practices, reflecting a practical shift in how intellectual property is managed within organizations. Because trade secret protection depends entirely on a company’s ability to maintain confidentiality, organizations must take on more internal responsibility through active oversight.
On a policy level, the shift toward trade secrets underscores how industry behavior can reshape the practical balance of the intellectual property system. Even without statutory change, heavier reliance on secrecy affects how much information ultimately enters the public sphere and how the patent system’s disclosure-based incentives function in practice. Further, the trend underscores how emerging technologies can test the limits of existing intellectual property regimes without necessarily prompting immediate legislative change. Over time, heavier reliance on secrecy may influence how information circulates within the industry. Many fields have long relied on a combination of publication, shared methods, and informal knowledge exchange to support technological development. As trade secret protection becomes more common, the movement of technical insights may become more limited, with fewer opportunities for open discussion or collective problem solving. This does not necessarily signal a decline in innovation, but it suggests that the ways innovation spreads could become more fragmented as confidentiality becomes a central feature of intellectual property management.