May 21, 2026
Artificial intelligence is becoming a bigger part of how people work, but some users are pushing that relationship to the extreme. A trend known as “tokenmaxxing” encourages people to maximize their use of AI tools.
For some, that philosophy comes with a high price tag. Speaking at the global tech conference Web Summit, software engineer Sigrid Jin suggested people should be using AI so much that the cost starts to rival one of their biggest monthly expenses: rent.
Advertisement
“If you want to know the future of AI, try to do tokenmaxxing,” Jin said, telling enthusiasts to encourage their friends “to try to spend [on] your AI pricing as much as you pay for your own house rent” in order to maximize their return on investment.
Here’s what tokenmaxxing is, and why treating AI spending like a monthly necessity could create problems for your budget.
What is tokenmaxxing?
As the term gains traction in AI circles, Pragati Awasthi, an Assistant Teaching Professor at Drexel University’s College of Computing & Informatics, explained the trend to Moneywise.
“At its simplest, a token is a small unit of text that AI models process roughly four characters, per OpenAI’s estimate,” she said. “Tokenmaxxing is the practice of maximizing how many tokens you consume, either to signal AI fluency, hit internal leaderboard metrics, or genuinely squeeze more output from AI tools.”
Subscription plans for AI products have usage limits that correspond to the number of tokens used. Tokens have a cost in the real world, as the servers used for these tasks have to be powered, cooled and maintained. Energy-intensive tasks like creating code or generating video use more tokens and cost more. So the competitive culture of “tokenmaxxing” is a bit counterintuitive: Workers who use more tokens cost their companies more money.
Nevertheless, the trend appears to be spreading inside tech workplaces, where some companies are beginning to closely track how employees use AI tools.
At Meta, an employee reportedly built an internal system to measure employee AI usage. Dubbed “Claudeonomics” — a nod to Anthropic’s large-language model, Claude — the system created a leaderboard ranking staff based on how many tokens they exchanged with AI models.
Advertisement
The tokenmaxxing concept “blew up” after news of Meta’s leaderboard leaked outside the company, and people started seeking out titles like “Token Legend,” Awasthi said.
Meta isn’t alone. Companies including OpenAI, Anthropic, Shopify and Sequoia Capital are also reportedly paying close attention to AI usage as the pressure to demonstrate AI productivity grows. At some tech firms, AI use has reportedly started showing up in performance reviews, rewarding workers who heavily embrace the tools and putting pressure on others to keep up.
The spending tied to these tools can also add up quickly. One Claude Code user at Anthropic reportedly incurred more than $150,000 in AI-related costs in just one month.
Must Read
- Dave Ramsey warns nearly 50% of Americans are making 1 big Social Security mistake — are you doing the same?
- Thanks to Jeff Bezos, you can now become a landlord for as little as $100 — and no, you don't have to deal with tenants or fix freezers. Here's how
- Robert Kiyosaki says this 1 asset will surge 400% in a year and begs investors not to miss this ‘explosion’
Join 250,000+ readers and get Moneywise’s best stories and exclusive interviews first — clear insights curated and delivered weekly. Subscribe now.
The return on a token investment
Part of the appeal of tokenmaxxing comes down to the return on investment. Jin argued that many people aren’t getting the full AI experience because they rely on free or lower-cost plans. Jin said users sticking with those versions could be missing out on more advanced tools available through premium subscriptions that can cost up to $200 a month.
When asked what kind of return people could expect from spending more on AI, Jin said: “running multiple businesses.” Tech-minded people are using AI to automate all sorts of tasks at work and at home, and the costs add up.
But experts say using more resource-intensive computing power through AI tools — what people in the business call “compute” — doesn’t automatically mean getting more value from it. Awasthi said both companies and workers are increasingly treating heavy AI use as a sign of productivity. Some businesses want measurable ways to show AI results, while younger workers may see frequent AI use as a way to signal they’re “AI-native.”
Advertisement
Still, Awasthi says activity and value aren’t necessarily the same thing.
“The problem is that token usage measures activity, not value. It’s like measuring a company’s revenue by the number of cold calls made,” Awasthi said.
Where should AI fit into your budget?
AI tools can save time and improve productivity, but that doesn’t mean they should come before core financial needs.
“If you can point to specific work that’s faster, better, or more valuable because of AI, the spend is justified,” Awasthi said. “For knowledge workers, freelancers, or anyone whose output directly affects their income, the ceiling is higher and the strongest case for spending more is a concrete return on output.”
For most people, AI subscriptions would likely fall into the “wants” category under the popular 50/30/20 budgeting rule, which recommends putting 50% of after-tax income toward needs, 30% toward discretionary spending and 20% toward savings and debt repayment.
Before paying for premium AI tools or diving deeper into “tokenmaxxing,” ask yourself: Is this actually saving me time? Am I using it enough to justify the cost? Could I get similar value from a free version?
The goal isn’t to avoid spending money on AI; it’s making sure the return shows up in your life and budget — not just your token count.
You May Also Like
- Turning 50 with $0 saved for retirement? Most people don’t realize they’re actually just entering their prime earning decade. Here are 6 ways to catch up fast
- Inside a $1B real estate fund offering access to thousands of income-producing rental properties — with flexible minimums starting at $10
- Vanguard’s outlook on U.S. stocks is raising alarm bells for retirees. Here’s why and how to protect yourself
- Here are 5 easy ways to own multiple properties like Bezos and Beyoncé. You can start with $10 (and no, you don’t have to manage a single thing)
Victoria Vesovski is a Toronto-based staff reporter at Moneywise covering personal finance, lifestyle and trending news. She holds degrees from the University of Toronto and New York University, and her work has appeared on platforms including Yahoo Finance, MSN Money and Apple News.
