In California, on March 6, 2026, Elon Musk's xAI sought to prevent the enforcement of a new law that mandates transparency about the training data used by AI companies. This law, known as Assembly Bill 2013, forces AI developers to disclose various important details concerning their datasets. Such disclosures involve specifics about the datasets used for training models, when they were collected, whether the data collection is ongoing, and if the datasets could contain protected data under copyright, trademark, or patent laws. xAI contended that this requirement infringes on trade secrets and would harm its competitive advantage in the AI marketplace. The company attempted to argue that revealing the sources of their training data could lead competitors, like OpenAI, to acquire critical datasets that would allow them to improve their models. However, in a ruling by US District Judge Jesus Bernal, it was determined that xAI failed to provide concrete evidence demonstrating that the disclosures mandated by the law would cause them significant harm. The judge noted that the claims made by xAI were overly vague, focusing on general concerns rather than specific instances where trade secrets could be compromised. Furthermore, the court acknowledged the interest of the public in understanding the training data behind AI models. The judge indicated that consumers would want to know if high-stakes data, such as medical information, was utilized in training AI models to determine the models’ reliability. This viewpoint challenges xAI's assertion that consumers do not care about such disclosures, as the court highlighted the necessary transparency for informed decisions in the AI marketplace. Additionally, xAI's legal team argued that California's law violated the Fifth Amendment rights related to intellectual property and trade secrets. Judge Bernal explained that while data can be considered trade secrets, xAI did not demonstrate that any of their datasets were unique or that their methods for cleaning and utilizing data differed meaningfully from those used by competitors. As a result, the court ruled that xAI was unlikely to succeed on those claims, with the emphasis being placed on the law's importance to foster public awareness regarding AI model capabilities and sources. Despite ongoing controversies surrounding xAI's chatbot, Grok, particularly regarding issues of hate speech and the generation of harmful materials, the judge clarified that the statute does not operate to suppress the company's opinions or influence model outputs.