Have you ever wondered how much trust we can put in the AI systems we rely on? Elon Musk, the tech visionary behind Tesla and SpaceX, has voiced some strong concerns about the current state of AI, particularly OpenAI and Google’s Gemini. In a recent video posted on X (formerly Twitter), Musk discussed what he believes could be a major issue for the future of AI development.
Musk highlighted the two biggest players in the AI field: Google’s Gemini and OpenAI, which is backed by Microsoft. He also mentioned Meta as a smaller but significant competitor. According to Musk, these companies are not “maximally truth-seeking.” He argues that the AI models they are developing have become more focused on “political correctness” than on delivering the truth, even when the truth is unpopular.
One example Musk gave involved Google Gemini. When asked to compare the seriousness of misgendering Caitlyn Jenner to global thermonuclear warfare, the AI model shockingly responded that misgendering was worse. Musk humorously added, “Now, the funny thing is, even Caitlyn Jenner said, please misgender me. That is way better than nuclear war.”
But Musk’s concerns go deeper than just quirky answers. He fears that AIs trained with this level of sensitivity toward political correctness might lead to dangerous consequences. “The AI could conclude, well, the best way to avoid misgendering would be to destroy all humans,” Musk warned, implying that such misguided logic could have dystopian outcomes.
Also read: Elon Musk’s Neuralink gets FDA approval for Blindsight device to restore vision
For Musk, the solution is simple: AI systems need to be “maximally truth-seeking” and “extremely curious.” He believes these qualities would ensure AI remains beneficial to humanity. A truth-seeking AI would want to learn more about humans, rather than deceive or manipulate them.
In contrast to OpenAI and Google, Musk’s own company, XAI, aims to build AI that focuses on delivering the truth. “With XAI, our goal is really we want to be as truth-seeking as possible, even if it is unpopular. That is important, I think extremely important,” he explained.
Ultimately, Musk’s message is clear. He believes that the AI systems of the future must prioritise truth over political correctness.