Enterprise AI adoption hinges on 90% cost cut, warns Palo Alto Networks CEO

Palo Alto Networks CEO Nikesh Arora has issued a stark warning that artificial intelligence token costs must fall dramatically before companies can embrace the technology at scale. According to a CNBC report, Arora said prices need to drop by roughly 90% for widespread enterprise adoption, reflecting growing frustration among business leaders with the current pricing models from major AI labs.

"We need to see the pricing for AI come down."

WhatfingerPalo Alto Networks CEO: We need to see the pricing for AI come down

Arora made the comment during an appearance on CNBC's 'Squawk on the Street,' where he also discussed concerns over high token costs and OpenAI's newest model, Whatfinger reported. The call for a massive price reduction echoes similar criticism from Palantir CEO Alex Karp, who has also pushed back against the token-based pricing systems used by companies like Anthropic and OpenAI, CNBC noted.

As token costs continue to climb, enterprises are increasingly turning to cheaper open-weight models as an alternative. CNBC reported that Chinese AI alternatives are rapidly catching up to American labs, offering comparable performance at a fraction of the price. This shift is happening even as investment in AI infrastructure surges to new highs, with companies pouring billions into hardware and data centers to support the technology.

Arora's remarks highlight a growing tension between AI developers and their enterprise customers. While the labs defend their pricing as necessary to cover massive compute costs, business buyers are pushing for more predictable and affordable access, especially as competing open-source models gain ground.