Amazon, Walmart, Cisco, and Meta are all curbing AI agent spending. The compute bill just arrived.

The Financial Times reported yesterday that Amazon, Walmart, Cisco, Uber, and Meta are all introducing usage caps on AI tools because the bills are out of control.

Cisco's chief product officer explained why: "For every human you might have 10, 100, or on the aggressive side 1,000 agents. They just keep working and that consumes a chunk of compute."

This is the structural break no one budgeted for.

Chatbots were cheap. Flat-rate subscriptions. Predictable costs. Agents are a different animal entirely. They chain multiple steps, run autonomously for hours, and consume 100x to 1,000x more tokens per interaction.

The math caught everyone off guard. Amazon scrapped its internal AI leaderboard after engineers gamed it by pointing agents at pointless tasks just to climb the rankings. Senior VP Dave Treadwell told staff: "Please don't use AI just for the sake of using AI."

Walmart's CTO said usage of their internal coding platform "really skyrocketed." Uber blew through its entire 2026 AI budget by April. Four months. Done.

Sam Altman acknowledged cost has become a "huge issue" this year. The issue never came up last year.

This is not a slowdown. It is a repricing.

Audit your agent deployment costs today. If your finance team cannot tell you the token cost per agent per workflow, you are flying blind. Every enterprise deploying agents at scale needs a FinOps function specifically for AI compute, or the next budget cycle will deliver a very unpleasant surprise.

SOURCE: https://www.ft.com/content/1d37cc08-e0aa-45a4-a45d-4ad282529314
VERIFIED: Financial Times (June 19, 2026), Business Insider (May 29, 2026), ETCIO/Times of India (June 20, 2026)
SIGNAL: The shift from chatbots to autonomous agents has broken enterprise AI budgets. Every company deploying agents needs real-time cost visibility or the next budget cycle will force painful cuts.