PANW CEO flags 90% AI price cut needed, challenging growth narrative

What happened

Palo Alto Networks CEO Nikesh Arora told CNBC that token costs for AI security must decline by up to 90% to drive enterprise adoption, a view paralleled by Palantir's Karp. This comes even as the company reported strong Q3 FY2026 demand: NGS ARR +60% y/y to $8.1B and RPO +36% to $18.4B, with raised FY2026 guidance. However, the DeepValue master report highlights extreme valuation (286x P/E, 120x EV/EBITDA) and GAAP operating losses due to $198M acquisition-related costs and $280M intangibles amortization, mostly from the CyberArk deal. Arora's pricing warning injects new risk that the AI security revenue stream may require sharp price compression, potentially undermining the platformization growth thesis and margin recovery hopes. Investors now must question whether the stock's elevated multiples can be sustained if AI monetization requires dramatic price cuts.

Implication

The CEO's statement suggests AI security pricing will compress sharply, pressuring revenue growth and margins—challenging the bull case of sustained 30%+ RPO expansion. The thesis now requires proof that volume can offset price cuts and that integration costs don't worsen. Reassess after FY2027 guidance and concrete evidence of AI deal economics.

Thesis delta

The CEO's pricing comment introduces a material risk that AI security monetization will require steep price cuts, contradicting the bull case of AI-driven premium pricing and accelerating adoption. This shifts the risk/reward balance downward, making the bear scenario more probable and the $280 entry potentially too optimistic. We reduce conviction in the bull case and emphasize waiting for clearer margin and pricing trajectory.