Social network X has recently updated its developer agreement to prevent third parties from using the platform’s content to train large language models. This change, which was spotted by TechCrunch, comes as part of a growing trend among major tech platforms to restrict AI data access. Specifically, the new clause under "Reverse Engineering and other Restrictions" states: "You shall not and you shall not attempt to (or allow others to) […] use the X API or X Content to fine-tune or train a foundation or frontier model." The timing of this policy shift is significant. In March, X was acquired by Elon Musk’s AI company, xAI, for $33 billion on paper. It is reasonable to infer that xAI does not want to provide competitors with unrestricted access to the social platform's rich data without proper compensation or agreements in place. This move aligns with xAI’s broader strategy of protecting valuable data resources that can be leveraged for AI development. The new restriction contrasts with previous policies. In 2023, X altered its privacy policy to allow the use of public data on its site for AI training. By last October, it had further relaxed restrictions to enable third parties to train their models. However, the recent update signals a strategic pivot toward tighter control over its data, which is a critical asset for training sophisticated AI systems. This shift is not unique to X. Other platforms have also tightened their policies regarding AI data access. Reddit, for instance, has implemented safeguards against AI crawlers and even filed a lawsuit against Anthropic last month. The lawsuit alleges that Anthropic’s AI system accessed Reddit more than 100,000 times since July 2024, violating the platform’s usage restrictions. Similarly, The Browser Company has added a clause to its AI-focused browser Dia’s terms of use to restrict AI training on its content. These moves reflect a broader industry concern about the ethical and commercial implications of AI data scraping. Platforms like X, Reddit, and The Browser Company are increasingly wary of their data being used to create competing products. By imposing such restrictions, they aim to protect their intellectual property and ensure that any commercial use of their data is properly authorized and compensated. The policy changes at X could also pave the way for more structured and profitable AI training deals. For example, Reddit recently secured a lucrative deal with Google, allowing the tech giant to train its AI models on Reddit’s data in exchange for financial and other compensations. X might be looking to follow a similar model, leveraging its vast trove of social media data to strike advantageous agreements with AI firms. Despite the restrictions for third parties, X’s own privacy policy still allows the platform to use user data for training its own AI models, including Grok. Users can opt out of this data sharing, but the default setting remains inclusive. This dual approach—restricting external access while maintaining internal use—highlights the platform’s commitment to both protecting its data and advancing its AI capabilities. Industry insiders view this policy shift as a necessary step to maintain a competitive edge in the rapidly evolving AI landscape. Alex Kipman, a prominent figure in AI development, commented, "This move by X is crucial for protecting their intellectual property and ensuring they can monetize their data assets effectively." He added, "Third-party scraping can lead to significant risks, both in terms of data security and potential legal battles, which these new policies help mitigate." Chris Nicholson, CEO of Skymind, another AI firm, noted, "While it might seem restrictive, these policies actually provide clarity and structure to the market. Companies can now focus on forming legitimate partnerships rather than engaging in ethically questionable data practices." In summary, X’s updated developer agreement is part of a broader trend in the tech industry to safeguard valuable data from unauthorized AI training. This move not only protects X’s assets but could also set the stage for more structured and financially beneficial data-sharing arrangements. As AI continues to advance, it is likely that more platforms will adopt similar measures to balance innovation with data protection. About X: X, previously known as Twitter, is a leading social media platform with over 400 million monthly active users. It has been at the forefront of real-time communication, influencing everything from political discourse to pop culture. With its acquisition by xAI, X is poised to integrate advanced AI technologies, aiming to enhance user experience and stay competitive in the tech sector. xAI, founded by Elon Musk, is a cutting-edge AI research organization focused on developing sophisticated language models and other AI applications. The acquisition of X demonstrates Musk’s commitment to combining social data with AI to drive technological and business advancements.