The key measure of leverage for AI-powered developers is no longer GPU utilization (FLOPs) but the volume of tokens processed by agents. Karpathy feels nervous when his token subscriptions are underutilized, indicating he's the bottleneck, not the system.

Paul Marshall suggests a proprietary, real-time indicator for market sentiment: his own firm's internal AI token consumption. A *deceleration* in the growth of this usage—not an outright decline—could signal that the excitement phase is peaking, providing an early warning to reduce risk before a market top.

After initial unrestricted spending led to budget overruns at companies like Uber, major enterprises are shifting focus. They are moving away from measuring raw AI usage (tokens) and toward implementing AI only for proven use cases with clear ROI, which may benefit cheaper, open-source models over expensive frontier ones.

While many teams track token usage as a primary AI ROI metric, Anthropic views it differently. High token usage doesn't correlate with high value, as it's easy to waste tokens. Instead, a token count of zero is the most important signal, as it clearly indicates someone is not using the AI at all.

In the current 'capability exploration' phase, companies incentivize developers to use as many AI tokens as possible. This serves as a visible, albeit inefficient, signal of AI adoption to management, prioritizing quantity over quality.

Categorize all AI token spend into three buckets: 'Tokens that Teach' (valuable experimentation), 'Tokens that Produce' (work output), and 'Tokens that Spin' (wasteful, idle processes). The optimal strategy is to aggressively eliminate 'spin' tokens, optimize 'produce' tokens, and fiercely protect the budget for 'teach' tokens to foster innovation.

The trend of companies like Uber and Meta capping employee AI usage, dubbed "token panic," does not signal a decline in overall AI demand. Instead, it marks a critical market shift towards prioritizing cost-effectiveness, creating a strong business imperative for more token-efficient models and applications.

A forward-looking business metric is emerging where capital allocation shifts from human labor to AI agent labor, measured in 'token spend.' Some tech-forward companies already have token budgets 20-50% higher than their human payrolls, signaling a fundamental change in how businesses will operate and measure productivity.

To gauge real AI integration, ignore official strategies and look at the data. Low per-employee token spend is a red flag indicating a lack of genuine, hands-on usage and curiosity, which are the real drivers of successful adoption.