Tokenmaxxing is out, valuemaxxing is in

August 18, 2026 | Source: Fast Company AI

Tags: Tesla, Meta, Uber, enterprise AI, AI spending, ROI

Tesla, Uber, and Meta are capping AI token spending and demanding ROI accountability after months of rewarding raw usage volume — a corporate reckoning that signals the end of the more tokens equals more progress era.

Details

The tokenmaxxing era — where companies encouraged heavy AI API and compute usage as a proxy for AI adoption — is closing. Fast Company reports that Tesla, Uber, and Meta are now imposing spending caps and shifting focus to valuemaxxing: AI deployments that demonstrably move business metrics rather than simply consume tokens. This shift was predictable. Enterprise AI spending grew through 2024-2025 largely on the assumption that usage equaled value. As CFOs started demanding ROI accounting, the mismatch between token volume and business outcomes became impossible to ignore. Companies that had enthusiastically given developer teams open API access are now adding governance layers, ROI thresholds, and usage controls. The pattern mirrors previous enterprise software waves (cloud spending rationalization, SaaS sprawl consolidation). The net effect for the AI infrastructure market may be a slowdown in raw token consumption growth even as the number of AI deployments continues rising. Content is from a Fast Company RSS lede — the full article with specific figures is behind a paywall.