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

This week, a topic that has been thriving in Silicon Valley burst into the spotlight: AI tokens as compensation. The idea is straightforward enough — instead of just giving engineers salaries, stocks, and bonuses, companies would also hand them a budget of AI tokens, the computational units that power tools like Claude, ChatGPT, and Gemini. Spend it to run agents, automate tasks, and manipulate code. The idea is that access to more computing makes engineers more productive, and that more productive engineers deserve more. It’s an investment in the person holding it, that’s the idea.

Nvidia’s leather-jacket-wearing CEO Jensen Huang seemed to capture everyone’s imagination when he floated the idea at the company’s annual GTC event earlier this week that engineers should get nearly half of his base salary back — in tokens. By his calculations, his senior staff might spend $250,000 a year on AI computing. He described it as a recruiting tool and predicted it would become a standard in Silicon Valley.

It is not entirely clear where the idea first originated. Tomasz Tunguz, a well-known Bay Area venture capitalist who runs Theory Ventures and focuses on AI, data, and SaaS startups — and whose writings on all things data have gained a loyal following over the years — was talking about this in mid-February, writing that tech startups were already adding inference costs “as a fourth element to engineering compensation.” Using data from compensation tracking site Levels.fyi, he determined the top quartile software engineer salary was $375,000. Add $100,000 in tokens, and you’re at $475,000 fully loaded – meaning roughly $1 out of every five dollars is now accounted for.

This is no coincidence. Agent AI is starting to take off, and the release of OpenClaw in late January dramatically accelerated the conversation. OpenClaw is an open source AI assistant designed to work continuously, switching between tasks, spawning sub-agents, and working on a to-do list while the user sleeps. It’s part of a broader shift toward “active” AI, meaning systems that don’t just respond to prompts, but take sequences of actions autonomously over time.

The practical result is that token consumption explodes. While someone writing an article might use 10,000 codes in an afternoon, an engineer managing a swarm of customers could blast millions in a single day — automatically, in the background, without typing a single word.

By the end of this week, The New York Times had put together a clever look at the so-called tokenmaxxing trend, finding that engineers at companies including Meta and OpenAI were competing on internal leaderboards that track token consumption. Generous token budgets have quietly become a standard job perk, much as dental insurance or free lunch once were, the newspaper reported. One Ericsson engineer in Stockholm told The Times that he probably spends more on Claude than he earns in salary, even though his employer foots the bill.

Tokens may become the fourth pillar of engineering compensation. But engineers may want to toe the line before embracing this as an outright win. More tokens may mean more power in the short term, but given how quickly things are evolving, that doesn’t necessarily mean more job security. For one thing, heavy allocation to cryptocurrencies comes with high expectations. If a company is effectively financing the costs of a second engineer’s computing for you, the implicit pressure is to produce at twice (or more) rates.

And there’s a more complex problem underneath: When a company’s premium spending per employee approaches or exceeds that employee’s salary, the financial logic of its headcount begins to look different from its finance team. If computation is doing the work, it becomes difficult to avoid the issue of how many humans need to coordinate.

Jamal Glenn, a Stanford MBA based on the East Coast and a former venture capitalist turned financial services CFO, points out that what may seem like a perk can be a smart way for companies to inflate the apparent value of a compensation package without increasing cash or stock — things that actually compound for an employee over time. Your nominal budget is not worth it. He can’t. It does not appear in subsequent offer negotiations in the same way as base salary or a stock grant. If companies succeed in normalizing tokens as pay, they may find it easier to hold cash steady while pointing to the increased computing allowance as evidence of investment in their employees.

This is a good deal for the company. Whether it’s a good deal for an engineer depends on questions that most engineers don’t have enough information to answer.