Intellectual property (IP) protection is crucial to preserving competitive advantages and incentivizing innovation. This is especially so in the rapidly evolving field of artificial intelligence (AI). While patents may offer IP protection for certain AI innovations, trade secret law may provide the most robust and flexible legal safeguard for AI technologies. Trade secret law offers distinct advantages over patent protection, particularly in areas where AI is still developing.

Trade Secret Protection

The Defend Trade Secrets Act (DTSA) defines trade secrets, in part, as “all forms and types of financial, business, scientific, technical, economic, or engineering information.” 18 U.S.C. § 1839(3). The owner of such information must take reasonable steps to keep the information secret, and the information must derive some economic value from not being generally known or ascertainable through proper means (that is to say, lawfully) by another person. 18 U.S.C. § 1839(3)(A-B). Reasonable steps may include measures such as using non-disclosure agreements and employed cybersecurity protection to prevent the information from being disclosed or misappropriated. With the DTSA’s broad definition, there is greater potential for information to meet trade secret subject matter eligibility. For AI technologies, trade secrets may cover a variety of eligible subject matter including learning models, proprietary algorithms, and training datasets.

Patent Protection

While patent protection provides strong IP rights against infringement, there are several challenges, particularly for AI technologies. First, the patent application process requires a detailed, public description of the invention, which may involve disclosing proprietary algorithms or AI models. The details of the invention become publicly available once the patent is granted, which presents a significant disadvantage for companies that aim to keep their AI technologies confidential.

Another drawback of patents in the context of AI is the time and expense of securing protection. AI technologies are changing every day, and it may be that a particular technology becomes outdated by the time a patent is granted. This puts companies, especially emerging companies, in the position of expending precious financial resources for less-than-guaranteed protection.

Benefits of Trade Secret Protection and AI

Trade secret law offers several advantages over patent protection for AI technologies:

1. Longer Period of Protection. While patents expire after 20 years, trade secret protection can last indefinitely as long as the secrecy is not broken.
2. No Disclosure. A key benefit of trade secret protection is that it allows companies to keep their AI innovations confidential. Unlike patents, which require public disclosure of the technology, trade secrets afford companies the opportunity to safeguard proprietary algorithms and data without having to reveal them to competitors.
3. Lower Costs for Protection. Securing patent protection requires considerable filing fees, legal costs, and time. In contrast, trade secrets are relatively inexpensive to protect as they do not require a formal filing with a government agency.
4. Potential Ease of Enforcement. Trade secret enforcement also may be more readily accessible as the owner can immediately pursue legal action if the information is misappropriated. The ability to enforce a trade secret is not dependent on administrative examination processes and procedures.

Challenges of Trade Secret Protection

Trade secrets, however, do come with their own limitations. For example, there is a risk of misappropriation. If an employee, partner, contractor, or other party with access to confidential information were to leak such information, the company stands to lose its competitive advantage. AI companies also face the risk of reverse engineering—the process by which another party analyzes and gains knowledge about the way a product or system works. Competitors may be able to reverse engineer the technology, making the once confidential information knowable, and therefore depriving a company of trade secret protection.

Conclusion

Companies offering AI technologies face rapid advancements in their field as well as competition in the marketplace. Trade secret protection may be more advantageous for companies due to lower costs and fewer administrative barriers to securing and enforcing trade secrets. Ultimately, the choice of IP protection depends on the specific business strategy and the nature of the AI technology involved, but for many companies, trade secret law can offer a comprehensive and flexible way to secure rights in their proprietary AI inventions.