Are AI tokens the new signing bonus or just a cost of doing business?

The concept of compensating engineers with AI tokens—computational units used to power AI tools like Claude, ChatGPT, and Gemini—has gained traction in Silicon Valley as a potential new form of employee benefit. Nvidia CEO Jensen Huang recently proposed that engineers receive AI tokens equivalent to about half their base salary, framing it as both a productivity booster and a recruiting advantage. This approach reflects a growing trend where AI compute costs are becoming an integral part of engineering compensation packages, alongside salary, equity, and bonuses. Venture capitalist Tomasz Tunguz highlighted this shift earlier in the year, noting that some startups are already including inference costs as a “fourth component” of compensation. Data suggests that top-tier engineers might receive token budgets worth $100,000 or more annually, making compute expenses a significant portion of total compensation. This development coincides with the rise of agentic AI—autonomous systems that perform complex sequences of tasks without direct human input—leading to a dramatic increase in token consumption. For example, engineers running multiple AI agents can use millions of tokens daily, far surpassing traditional usage patterns. The New York Times has reported on this “tokenmaxxing” trend, revealing that companies like Meta and OpenAI track internal leaderboards based on token usage, turning generous token allowances into a coveted job perk. Some engineers, such as one at Ericsson in Stockholm, reportedly consume more AI compute than their salary, with employers covering the costs. While this signals a potential shift in how tech talent is compensated, there are concerns about the implications. More tokens may not necessarily translate into straightforward gains for engineers, raising questions about the sustainability and fairness of token-based compensation as it becomes more widespread in the tech industry.