In a significant policy shift, X, the company associated with Elon Musk, has amended its privacy policy to allow the use of user-generated information for training its artificial intelligence (AI) models. The updated privacy policy is set to take effect on September 29, introducing provisions that permit the utilization of publicly available data for machine learning and AI purposes.
This development came to light when Elon Musk, the entrepreneur behind X and SpaceX, responded to a post about the policy change on the platform, stating that only publicly available information would be used, and no private or direct messages would be accessed.
NEWS: X’s updated Privacy Policy has added a mention of using X data to train AI models 👀 pic.twitter.com/Bn5YyoeHjq
— X News Daily (@xDaily) August 31, 2023
In a live audio session hosted on X, formerly known as Twitter, in July, Musk had previously disclosed that his AI startup, xAI, intended to utilize publicly available data from the social media platform to improve its AI models. Despite reaching out to X for comments, Insider received no immediate response regarding the specifics of how X plans to employ the information or which AI models would be involved.
Also Read: Rapid Advancements in AI Pose Similar Risks as Social Media
Elon Musk has been vocal about his concerns regarding other companies training AI models using Twitter data. Earlier this year, he threatened legal action against Microsoft for what he deemed “trained illegally using Twitter data.”
They trained illegally using Twitter data. Lawsuit time.
— Elon Musk (@elonmusk) April 19, 2023
Musk’s xAI venture, aimed at using artificial intelligence to gain a deeper understanding of the universe’s true nature, was announced earlier this year. This launch occurred despite Musk’s prior warnings about the potential “destruction of civilization” stemming from AI and his signing of a letter advocating a pause in the training of more powerful AI systems, such as OpenAI’s GPT-4.
It is worth noting that Musk’s involvement in the AI sector predates xAI, as he was among the early backers of OpenAI. However, he departed the company’s board of directors in 2018, though he emphasized his pivotal role in the company’s inception, even contributing to its naming.
Moreover, Elon Musk has recently criticized ChatGPT, a conversational AI model, accusing it of being “The danger of training AI to be woke – in other words, lie – is deadly”
The danger of training AI to be woke – in other words, lie – is deadly
— Elon Musk (@elonmusk) December 16, 2022
The reliance on publicly available data for training AI models is common in the AI industry, with the technology underpinning generative AI products like ChatGPT demanding substantial amounts of data. Historically, data was sourced from publicly accessible web platforms such as Reddit and Wikipedia.
Also Read: Revolutionizing Learning: The Emergence of AI Quiz Generators in the Modern Classroom
Nevertheless, some companies are now exploring opportunities to monetize or restrict access to such data. For instance, Reddit announced plans in April to charge AI companies for data access, while entities like The New York Times and Amazon have begun blocking OpenAI’s web crawler.
Social media platforms, teeming with user-generated content, are a valuable resource for AI companies seeking fresh data sources for training.
In a related development, Meta, as it progresses with its generative AI models, has recently unveiled a feature that enables Facebook users to choose not to share their data for AI model training purposes. However, this measure has limitations, and complete removal of user data from their AI training databases is not guaranteed, as reported by Gizmodo.
As X moves forward with its revised privacy policy, it remains to be seen how this change will impact the AI landscape and user data privacy concerns, as Musk’s ventures continue to push the boundaries of artificial intelligence.
Leave a Reply