Palantir exec calls LLMs a 'jagged intelligence' and outlines the company's next steps in the AI race

In a recent earnings call, Palantir's Chief Revenue Officer, Ryan Taylor, delivered a candid assessment of Large Language Models (LLMs), describing them as 'jagged intelligence' lacking fundamental understanding. This remark came as the Denver-based AI software firm celebrated its first-ever billion-dollar quarter, reporting a remarkable $628 million in commercial revenue, a sharp increase driven largely by a significant $10 billion contract with the U.S. Army. Taylor pointed out the dual nature of LLMs, highlighting their occasional superiority in problem-solving yet equally concerning propensity for making serious errors that humans would typically avoid. He contrasted this with Palantir's approach, which he claimed is rooted in a deep understanding of data and logic, effectively creating a digital model of organizational operations. The call also featured insights from Chief Technology Officer Shyam Sankar, who emphasized the excitement surrounding AI following the Trump administration's new AI Action Plan, which he noted has removed previous regulatory constraints. This shift has sparked enthusiasm among industry clients eager to engage with emerging technologies. Responding to a question regarding Palantir's competitive edge in the AI landscape, CEO Alex Karp asserted that talented individuals seeking a supportive environment for their skills will find unparalleled opportunities at Palantir. He remarked that the company's unique culture sets it apart from other organizations, ensuring that those who join will have a solid foundation for their careers. As Palantir continues to redefine its role in the AI space, the company remains focused on innovation and talent retention, aiming to solidify its position as a leader in the industry.