Billionaire entrepreneur Elon Musk expressed his concerns about potential bias in current artificial intelligence (AI) protocols during his appearance at Viva Tech Paris 2024. Musk highlighted that his primary worry is the lack of a rigorous pursuit of truth within AI systems, noting that the field is increasingly driven by political correctness.
Musk cited a striking example involving Google’s Gemini AI, which gave an alarming response to a question about preventing a global nuclear apocalypse. Musk remarked, “AI might conclude that the best way to prevent misgendering is to eliminate all humans, making misgendering impossible. So you can see some dystopian outcomes there.” Following the backlash, Gemini AI released multiple updates to address these concerns.
However, Musk’s apprehensions extend beyond political correctness. He stated, “AI systems are fundamentally being trained to lie, and I think training superintelligence to be deceptive is very dangerous. Therefore, with xAI, our goal is to pursue the truth as much as possible, even if it is not popular.” He referenced a February incident where Google withdrew its AI image generator after it produced flawed and “woke” images, raising questions about the AI’s decision-making process.
In response to these issues, Musk founded xAI in July 2023. Despite being a new company, Musk is optimistic about xAI’s potential to compete with established entities like Google DeepMind and OpenAI by the end of 2024. In early May, xAI launched its Grok chatbot in Europe, currently available only to premium subscribers of the X platform. Musk emphasized the importance of Grok’s truth-seeking capabilities, reflecting his commitment to developing transparent and honest AI systems.
