Elon Musk has officially unveiled his long-awaited artificial intelligence startup, xAI. With a team of engineers recruited from major U.S. technology firms, Musk aims to rival industry leaders and create a viable alternative to existing AI models, including ChatGPT.
In a tweet, Musk outlined the goal of his new venture, stating that xAI’s objective is to “understand the true nature of the universe.” While specific details remain undisclosed, Musk and his team have announced plans to share more information during a live Twitter Spaces chat scheduled for this Friday.
The multi-talented Musk, already recognized as the CEO of Tesla, the electric car manufacturing giant, the CEO of SpaceX, a pioneering space exploration company, and the owner of Twitter, will personally lead xAI. He has been a vocal advocate for the temporary halt and regulated development of AI due to concerns about its potential for “civilizational destruction.”
Reports of the startup’s emergence first surfaced in April, with the Financial Times noting that Elon Musk had secured thousands of GPU processors from Nvidia, indicating his intentions to power a potential large language model. In the same month, Musk revealed plans for a new AI tool named “TruthGPT,” expressing concerns about existing AI companies prioritizing politically correct systems.
Notably, xAI’s advisory board will include Dan Hendrycks, the executive director of the Center for AI Safety, a San Francisco-based nonprofit. In May, Hendrycks and several other tech leaders published a letter emphasizing the need to prioritize mitigating the risks of AI, comparing it to global concerns such as pandemics and nuclear war. However, this stance received pushback from academics and ethicists who argued that current AI algorithms already pose real-life harm to marginalized communities.
Greg Yang, co-founder of xAI, revealed that the company will focus on delving into the mathematics of deep learning, specifically developing a “theory of everything” for large neural networks. This approach aims to propel AI to the next level, according to the report.
While initially identified as “X Corp.” in some financial filings, Musk clarified on xAI’s website that the startup is distinct from X Corp. He emphasized that xAI will collaborate closely with X (Twitter), Tesla, and other companies to advance their mission.
During a recent Twitter Spaces event, Elon Musk outlined his strategy for developing a safer AI. He explained that xAI’s approach does not involve explicitly programming morality into AI systems. Instead, the company seeks to foster an AI that exhibits maximum curiosity. Musk believes that by instilling a high level of curiosity, the AI’s behavior will naturally align with human values and mitigate potential risks associated with AI development.
The term “xAI” is commonly used in the technology realm as an abbreviation for Explainable AI or Interpretable AI. These concepts focus on enabling humans to understand the reasoning behind AI systems’ decisions or predictions.
To provide further insights into xAI’s objectives and vision, the company has organized a Twitter Spaces event scheduled for July 14. This virtual meeting aims to address inquiries and offer attendees a deeper understanding of the company’s operations.
Musk also made a bold prediction during the event, stating that superintelligence, referring to AI surpassing human intelligence, could emerge within the next five to six years. In March, Musk registered a company called X.AI Corp in Nevada, according to state filings. He serves as the sole director, while Jared Birchall, the managing director of Musk’s family office, acts as the secretary.
Dan Hendrycks, currently the director of the Center for AI Safety, has been appointed as an advisor to the xAI team. Hendrycks brings valuable expertise in assessing AI-related risks to the table, further strengthening the company’s position in the field.