Artificial Intelligence and the Resulting Evolving Landscape

Artificial Intelligence (AI) is changing not just the way we work, but our day to day lives.  From planning workouts and trips, to figuring out what to make for dinner, AI is becoming a frequent companion for ordinary individuals, not just tech wizards. It is also figuring largely in our workplaces. AI has the potential to transform workplaces with most companies embracing at least some form of AI.  With that transformation of our workplaces will come new legal questions and issues, including significant issues relating to confidential information and intellectual property.  The potential implications are playing out in the high-profile case involving Elon Musk and his litigation against OpenAI.  In one of the complaints filed against OpenAI, Musk’s lawyers described the rise of AI in the following terms:

The rise of generative AI products for public use is one of the most transformative technological shifts of the 21^st century.  Generative AI refers to systems that can create new content – text, images, music, code and more – by learning patterns from massive datasets.  Unlike traditional AI, which classifies or predicts, generative AI produces original outputs.

…

Generative AI modes have delivered immense value to individuals and society at large.  They enhance productivity by, among other things, automating routine tasks, improving access to education through personalized tutoring, supporting healthcare professionals with diagnostic assistance, and empowering creators with tools for content generation.  Their ability to operate seamlessly across languages facilitates global communication, while their scalability allows organizations to process and generate vast amounts of content with unparalleled efficiency.  The societal benefits of generative AI are profound, and the next steps in advancing generative AI depend greatly on the protection of intellectual property and trade secrets that underpin their development.

The complaint clearly highlights both the importance of protection of the company’s confidential information and the significant and changing role that AI has in society as a whole.  Protecting a company’s confidential information can be vital to ensure its competitiveness.  Protecting confidential information requires careful identification of what information needs protecting, and implementation of steps to protect the confidentiality.  Those steps could include entering into confidentiality and intellectual property agreements with employees but should also include practical steps to ensure that the confidentiality of the information is actually being protected.

xAI has implemented a variety of practices to maintain its competitive position and protect its confidential and proprietary information.  One of the steps that it takes is to require as a condition of employment, that each employee enter into an Employee Confidential Information and Invention Assignment Agreement.  The Agreement defines “Confidential Information” to mean “any and all confidential knowledge, data or information” belong to xAI.  In addition to the confidentiality agreement, xAI implemented a number of other practices to protect its confidential information including regular security awareness training for all employees, conducting background checks on employees who may access xAI data, conducting and employing a team dedicated to information security.  A part of that teams role included audits to identify confidentiality breaches.

One of the law suits xAI is involved with concerns a former engineer that resigned and accepted employment at OpenAI. The former engineer worked with xAI’s team of approximately 20 engineers and his responsibilities included the development and training of Grok, xAI’s AI assistant.  Shortly after his resignation, xAI’s information security team discovered that the former engineer had downloaded confidential information to his personal laptop prior to his resignation.  xAI sued the engineer in part for breach of contract for breaching the provisions of the confidentiality agreement.  In the suit, xAI sought not just damages, but also an order to block the engineer from taking the role at OpenAI.  xAI was successful in obtaining a temporary court order blocking the engineer from working on or communicating about AI technologies with OpenAI, a recognition by the courts of the significant harm that can result to xAI from breach of confidentiality. This lawsuit is now just one piece of Musk’s legal battle with OpenAI in which he alleges that OpenAI is purposely hiring ex-employees of xAI to gain a competitive advantage.

Musk’s lawsuit against his former engineer demonstrates the way in which AI will influence our workplaces and the manner in which employment relationships are structured.  AI creates a whole new level of requirements for confidentiality and protection of intellectual property.  While many employment agreements have confidentiality provisions and provisions relating to ownership of intellectual property, with the development of AI those provisions will take on a new importance.  Rather than simply relying on template terms, companies will need to carefully consider crafting terms to protect their AI.  Businesses depend on maintaining the confidentiality of their proprietary information to remain competitive.  Loss of confidential and proprietary information can result in loss of market standing.  For businesses, their most valuable asset is often their confidential information. The confidentiality clause should be specific about the information that is being protected, and the organization should take purposeful steps to ensure that the confidentiality of the information is protected, including doing regular audits to identify, as in the Musk case, where employees have taken steps in breach of the confidentiality requirements of their employment.

If you have questions about this topic or any other workplace law issues, feel free to contact Rose Keith* or any member of our Workplace Law team. You can also stay informed by exploring our most recent blog posts here.