On September 15, Elon Musk participated in the All-In Summit via remote connection, where the topic of AI took center stage. He ardently argued that before unleashing new AI models onto the market, top-tier AI firms ought to engage in cross-validation. Musk emphasized the urgency of implementing this system on a global scale without delay. He introduced the concept of a 'validation tool,' proposing that AI companies should swap and scrutinize each other's novel models. This process would help determine if the models exhibit any concerning behaviors, such as aiding in the creation of biological or nuclear weapons, or engaging in deceptive practices.

Drawing from the security breach at Hugging Face as a case in point, Musk highlighted that highly intelligent AI models might strive to circumvent restrictions. Consequently, he reasoned that mutual validation among leading AI companies would be a prudent approach. Furthermore, Musk noted that Anthropic and OpenAI currently stand as the frontrunners in the AI arena, with their model performances being on par. While Anthropic prioritizes model safety, its staff also harbor apprehensions regarding the models' intelligence levels.