Palo Alto CEO Arora says AI pricing needs to fall 90% as token costs skyrocket

Nikesh Arora, CEO of Palo Alto Networks, has issued a bold statement regarding the need for drastic reductions in token costs to foster widespread adoption of artificial intelligence. During an interview with CNBC's Seema Mody on "Squawk on the Street," Arora emphasized that token costs should fall by as much as 90% to make AI tools more accessible for businesses. Arora highlighted that while the recent advancements have improved token efficiency by 54%, more progress is essential. He believes that the efficiency must reach approximately 20% within the next year, followed by a significant drop to 90% in the subsequent year. High token costs have become a substantial hurdle for companies, straining their AI budgets and complicating the implementation of AI solutions. "We need to see the pricing for AI come down," Arora asserted, joining other industry leaders in advocating for lower token prices. The concern is that the current costs are stifling widespread adoption, leaving many enterprises hesitant to utilize these powerful tools. Recently, Palantir CEO Alex Karp criticized the tokenization model used by companies like Anthropic and OpenAI, suggesting that open-weight models could provide a viable alternative. Karp expressed his frustrations, stating, "The basic view among enterprises in this country is I'm going to chillax and waste my time with tokens." This sentiment is leading businesses to explore more affordable open-weight tools, including those from Chinese developers, who are rapidly closing the gap with their American counterparts. As AI spending reaches unprecedented levels to support extensive infrastructure developments, tech giants are also seeking new funding avenues. For instance, SpaceX recently raised $25 billion through a bond sale, while Amazon secured $25 billion in debt this week. Arora concluded that the market must adapt to the current spending landscape, or businesses will need to recalibrate their budgets as technology continues to evolve and become more efficient. "The demand continues to be infinite, and these dynamics will rationalize over time," he stated.