Researchers have developed a method called Salt that uses generative AI to verify machine code at scale. This approach inverts the traditional relationship between verification and productivity, making verification not only economical but essential for large-scale autonomous machine work. The Salt method relies on a proof kernel that prevents hallucinated proofs from passing through, allowing mathematical claims to be checked as kernel-checked artifacts. Human attention is reserved for high-level statements and designs, freeing up time for more complex tasks. --- Why it matters: This matters because it enables one person to direct a small fleet of AI agents at scale without requiring human review or intervention in low-level details. This has significant implications for productivity and the potential for autonomous machine work to be safely directed by humans. Source: https://arxiv.org/abs/2608.21356