Tokenmaxxing and the Corporate AI Pullback

Summary

Down to Business English

Episode Summary: Tokenmaxxing and the Corporate AI Pullback

Date: June 13, 2026

Hosts: Skip Montreux & Des Morgan

---

Episode Overview

This episode explores the phenomenon of "token maxing"—the overconsumption of AI computing resources within corporations, leading to massive, unexpected costs and prompting a shift in how companies approach AI integration. The hosts also discuss the emerging “AI pullback,” where companies are rolling back or strictly limiting AI spending, and what this means for the future of business, productivity, and the AI industry.

The episode balances business insights with accessible language, focusing on learners of English as a second or foreign language (ESL/EFL). The hosts integrate real-world examples with practical vocabulary, idioms, and phrases useful in international business contexts.

---

Key Discussion Points & Insights

Introduction: A Personal Example of AI Overuse

• Des shares his AI app dilemma: Des explains he upgraded to the pro version of a calendar app for $120/year because of its AI features, including an “AI efficiency score.”
  • “I'm currently in the top 10% of users. I really want to keep my high score.” (Des, 01:21)
• Skip connects this to bigger corporate issues: Sees Des’s behavior as “a brand new corporate crisis called token maxing.” (Skip, 01:50)

What is Token Maxing?

• Definition: In commercial AI, companies are billed based on the number of tokens (basic units of data) processed. Token maxing happens when employees or systems use far more tokens than necessary, either through inefficiency or deliberate overuse to boost their "productivity" stats.
  • “Developers waste millions of these tokens. Sometimes they do it intentionally to pad their productivity stats…” (Skip, 02:08)
• Major Example: Uber burned through its entire annual AI budget in four months due to token maxing. (Skip, 02:55)

Agentic AI and the Consumption Crisis

• Agentic AI Explained: Unlike basic AI tools (which perform single tasks on command), agentic AI can undertake complex projects autonomously, cycling through tasks until a goal is met.
  • “Instead of doing one single task, it acts as an independent worker. You give it a goal, and it goes to work.” (Skip, 04:53)
• Resource Consumption: Agentic AI can use up to 1,000 times more tokens than standard AI prompts, especially when operating unsupervised in “continuous loops.” (Skip, 06:03)
• Cost Implications: This rampant token usage led Uber and other companies to enforce strict spending caps, such as a $1,500 monthly limit per developer. (Skip, 06:31)

Broader Corporate Impact

• Not Just Developers: Token maxing isn’t limited to tech roles. Any department handling large volumes of digital information—like legal, HR, finance—could face similar overages.
  • “Coders are just the canary in the coal mine. Uber is the headline example. But the bigger story is that this problem could follow agentic AI into any department…” (Skip, 07:03)

Corporate Responses to AI Overuse

• Amazon: Addressed competitive “token leaderboard” usage culture by removing such leaderboards entirely. (Skip, 08:30)
• Microsoft: Canceled most internal licenses for costly “Claude Code” agentic AI after budgets were drained. Forced teams to use less expensive alternatives. (Skip, 08:51)

Economic Reality of AI Adoption

• ROI Questioned: The anticipated cost savings from replacing humans with AI have not been realized if indirect costs—like supervising AI (“babysitting tax”) and runaway resource consumption—are included.
  • “Because the AI runs in these massive, endless loops, the actual computing costs...are astronomically higher than anyone predicted.” (Skip, 09:33)
  • “When you combine that babysitting tax with unexpected token costs, the financial math falls apart.” (Skip, 10:22)

Shifting from Token Usage to Inference Yield

• New Measurement Focus: Companies are moving away from simply tracking how much AI is used, and instead are measuring its output—how much useful work was actually accomplished per token spent.
  • “Developers no longer get a high score just for using AI. In fact, if a developer burns through millions of tokens just to produce one or two lines of useful code, their inference yield is terrible and they are considered a financial liability.” (Skip, 11:21)
  • “It’s like measuring a writer by the quality of the books they publish, rather than just the number of words they write.” (Des, 11:47)

Implications for AI Providers & the Market

• Growth at Risk: As major corporations clamp down on AI usage and spending, the revenue forecasts underpinning AI companies’ IPO ambitions face a “harsh reality check.”
  • “If the token maxing party is over, these massive IPO ambitions and trillion dollar valuation hopes could be in serious trouble.” (Skip, 13:14)

---

Notable Quotes & Memorable Moments

• On the addictive nature of AI usage:
  “You are intentionally using unnecessary computing power just to get a high score on an AI digital leaderboard?” (Skip, 01:34)
• On shifting corporate KPIs:
  “The focus is no longer on using more AI. It’s getting better results from the AI you use.” (Des, 11:13)
• On economic limits of tech enthusiasm:
  “The roi, or the return on investment isn't as high as they had originally thought.” (Des, 10:00)
• On market correction:
  “The tech market is in for a very harsh reality check.” (Skip, 13:14)

---

Important Timestamps & Segments

• [01:50] – Introduction of “token maxing” as a concept.
• [02:55] – Case study: Uber’s entire AI budget used up in 4 months.
• [04:53] – Explanation of “agentic AI.”
• [06:03] – Agentic AI can use 1,000x tokens per task.
• [06:31] – Implementation of token budgets at Uber.
• [07:03] – Expansion of the problem to non-technical departments.
• [08:30] – Amazon and Microsoft’s responses to token maxing.
• [10:22] – “Babysitting tax” – the managerial cost of supervising AI.
• [11:00] – Introduction of “inference yield” as the new metric.
• [13:14] – Discussion on how corporate AI pullback affects the market.

---

D2B Vocabulary Segment & Business English Takeaways

Key Terms Introduced (with explanations and examples)

• Color code ([13:59] onward): Organize information by color for categories.
• Burn through / Blow through ([15:39]): Spend or use up resources rapidly.
• A canary in the coal mine ([17:54]): An early warning indicator of larger problems.
• Reality check ([19:58]): A moment when unrealistic expectations are confronted by facts.

Quotes:

• “A canary in the coal mine is an early warning sign that a bigger danger or problem may be coming.” (Skip, 17:54)
• “A reality check is a moment when you are forced to stop, look at a situation honestly...” (Des, 19:58)

---

Conclusion

The episode closes by emphasizing how the story reflects a broader business reality: “What starts as a story about employees using too much AI quickly turns into a much bigger question about budgets, productivity, and whether companies are getting real value from these AI tools.” (Skip, 22:40)

Des muses on whether his own AI tool is truly worth it, encapsulating a dilemma now facing businesses worldwide.

---

For Listeners

This episode is especially useful for:

• Understanding key business English vocabulary and phrases related to AI, technology spending, and organizational change.
• Gaining insight into real, current business dilemmas and how they are being handled in large corporations.
• Reflecting on one’s own use (and value) of AI and digital tools in a work context.

---

For further vocabulary practice, bonus content is announced for subscribers and D2B members.

Transcript

[Narrator]

From Tokyo, Japan and Changsha, China, this is down to Business English with your hosts, Skip Montreux and Des Morgan.

[Skip Montreux]

Hello, Des. How's it going?

[Des Morgan]

Hi, Skip. I'm doing well, thanks. But I have to admit, I'm feeling a little guilty about a recent purchase I made.

[Skip Montreux]

Oh, what did you buy?

[Des Morgan]

I just upgraded to the pro version of my calendar app, and it cost $120 a year.

[Skip Montreux]

120 bucks for a calendar app? Well, why in the world do you need a pro version of a calendar app, Des?

[Des Morgan]

Well, it uses three different AI programs running constantly in the background. They read all my emails, analyze my tone, and then color code my meetings based on their importance.

[Skip Montreux]

Des, you are a teacher. You have the same classes and the same faculty meetings every single week. Your schedule is highly predictable. You do not have anywhere near enough meetings to justify spending $120 a year on an AI calendar app.

[Des Morgan]

But the app gives me an AI efficiency score based on how much the AI works for me. And I'm currently in the top 10% of users. I really want to keep my high score.

[Skip Montreux]

So let me get this straight. You are intentionally using unnecessary computing power just to get a high score on an AI digital leaderboard?

[Des Morgan]

Exactly. It makes me feel productive.

[Skip Montreux]

Well, as silly as that sounds, Des, you are perfectly illustrating a brand new corporate crisis called token maxing.

[Des Morgan]

Token maxing? What's that?

[Skip Montreux]

In the AI world, a token is a basic piece of information. When a company uses commercial AI, they are billed based on the total number of tokens their employees consume. Token maxing happens when developers waste millions of these tokens. Sometimes they do it intentionally to pad their productivity stats, but often it is just very inefficient coding where the AI is left running unnecessary tasks in the background.

[Des Morgan]

So they're playing the exact same game I am, but with the company's money?

[Skip Montreux]

Well, not exactly, but pretty close. And the implications are massive. For example, because of token maxing, the ride sharing company Uber just burned through its entire annual AI budget in only four months.

[Des Morgan]

That sounds like a serious crisis for the tech sector.

[Skip Montreux]

It is.

[Des Morgan]

I want to learn more.

[Skip Montreux]

Then let's do it. Let's get D2B down to business. With token maxing and the corporate AI pullback.

[Des Morgan]

I'm curious. You said Uber blew through its entire 12 month AI budget in just four months?

[Skip Montreux]

That's right. And that forced the company to put emergency limits in place.

[Des Morgan]

How did they manage to spend so much money that fast?

[Skip Montreux]

Good question. It comes down to how their software engineers were using AI. Specifically a New approach known as agentic AI.

[Des Morgan]

Hold on a minute. What do their software engineers even need AI for? Uber already has a working app. You push a button and a car shows up.

[Skip Montreux]

Well, it might work seamlessly on your phone, but a global app like Uber's is always evolving. Engineers are constantly rewriting the code, updating algorithms, maintaining massive databases, revising their user interface.

[Des Morgan]

Okay, so there's a lot of tedious behind the scenes maintenance work.

[Skip Montreux]

Exactly. And this is where agentic AI comes in.

[Des Morgan]

And what exactly do you mean by agentic AI?

[Skip Montreux]

Well, most people use AI like a search engine. You ask a question and you get an answer. Or you prompt the AI to do a task, like create a spreadsheet or write an email, and the AI does that for you.

[Des Morgan]

Or create crazy videos of your cat breaking into your bedroom late at night, playing a guitar and serenading you with a Mexican ballad.

[Skip Montreux]

Yes, there is some of that AI slop too. But agentic AI is different. Instead of doing one single task, it acts as an independent worker. You give it a goal, and it goes to work accomplishing it for you, using whatever tools it thinks are necessary and making decisions on its own to accomplish the task.

[Des Morgan]

So instead of just asking AI to draft an email, you give it an entire project to work on from start to finish.

[Skip Montreux]

Exactly. Uber engineers were giving agentic AI tools broad goals, like reviewing code or updating old databases and other similar tasks.

[Des Morgan]

And the AI just does all that work automatically?

[Skip Montreux]

That's right. The AI works in continuous loops, searching through millions of lines of old code, writing updates, testing them, and correcting its own errors over and over again in the background.

[Des Morgan]

I see. And all that background reading and writing consumes tokens.

[Skip Montreux]

A massive amount of tokens. In fact, industry data shows that an agentic AI can consume up to 1000 times more tokens than a standard AI prompt just to complete a single task.

[Des Morgan]

A thousand times more. No wonder developers are burning through corporate budgets. So what exactly did Uber do when they realized all the money was gone?

[Skip Montreux]

They immediately hit the brakes and instituted an emergency company policy. They placed a strict $1,500 monthly token limit on every single developer. Once an Engineer hit that $1,500 cap, their access to premium AI coding tools was cut off for the rest of the month.

[Des Morgan]

A strict $1,500 limit. Interesting, but skip. This isn't just a problem for software engineers, is it?

[Skip Montreux]

You're right. Not at all. Coders are just the canary in the coal mine. Uber is the headline example. But the bigger story is that this problem could follow agentic AI into any department that handles a large amount of information. Legal, human resources, marketing, finance, and many other departments.

[Des Morgan]

Can you give us an example outside of tech?

[Skip Montreux]

Sure. Well, imagine a law firm. Instead of paying a human employee to spend a week reading through old contracts, they unleash an agentic AI. You give it a goal, and the AI reads 10,000 pages of legal text, cross references corporate policies, and drafts a final report. All in an hour or two.

[Des Morgan]

Much faster than a human could, obviously.

[Skip Montreux]

Sure, but it brings the exact same token maxing problem. If that legal AI gets confused by a loophole, it might run in a continuous loop, reading the same files over and over, burning through tokens.

[Des Morgan]

Which brings us back to the crackdown on token usage. Are other companies stepping in to stop this waste?

[Skip Montreux]

Let's look at two other responses. Amazon focused on employee behavior. Microsoft focused on software access.

[Des Morgan]

Okay, start with Amazon. What did they do?

[Skip Montreux]

Amazon realized their employees were treating AI usage like a video game leaderboard, competing to see who could use the most tokens. So the company completely shut that leaderboard down.

[Des Morgan]

And what about Microsoft? They were one of the first to jump on the AI bandwagon. Are they doing anything about this?

[Skip Montreux]

Absolutely. In fact, Microsoft just canceled the majority of its internal licenses for Claude Code, a premium AI tool. Their developers were running so many agentic AI loops that the token cost blew through the company's annual budget in a matter of months.

[Des Morgan]

That's not good.

[Skip Montreux]

Microsoft is now forcing its teams to switch to less expensive AI tools to stop that bleeding.

[Des Morgan]

But hold on, Skip. Even if A company caps AI spending at $1,500 a month, isn't that still much cheaper than paying a human paralegal or software engineer's salary?

[Skip Montreux]

That was the original theory. The promise of AI was that it would be incredibly cheap and could replace costly human workers. But the cost of agentic AI is proving that to be wrong. Because the AI runs in these massive, endless loops, the actual computing costs involved in using AI are astronomically higher than anyone predicted.

[Des Morgan]

So the roi, or the return on investment isn't as high as they had originally thought.

[Skip Montreux]

It is highly questionable. Companies are realizing you cannot just lay off a human and plug in an AI. When you combine these massive, unexpected token bills with the babysitting tax.

[Des Morgan]

Sorry, babysitting tax?

[Skip Montreux]

Yes, where you still have to pay a senior manager to fix the AI's mistakes. When you combine that babysitting tax with unexpected token costs, the financial math falls apart.

[Des Morgan]

So in the end, replacing human employees with AI might not actually be worth It.

[Skip Montreux]

And that realization is why many businesses are changing their AI strategy. They are moving away from measuring raw token usage and shifting to a concept called inference yield.

[Des Morgan]

Inference yield. Okay, break that down for us.

[Skip Montreux]

Companies are moving from a simple question, how much AI did employees use? To a better question, how much useful work did that AI actually produce?

[Des Morgan]

So the focus is no longer on using more AI. It's getting better results from the AI you use.

[Skip Montreux]

In Amazon's case, it means tracking efficiency, specifically the ratio of tokens used to useful code published. Developers no longer get a high score just for using AI. In fact, if a developer burns through millions of tokens just to produce one or two lines of useful code, their inference yield is terrible and they are considered a financial liability.

[Des Morgan]

Well, that does make a lot of sense. It's like measuring a writer by the quality of the books they publish, rather than just the number of words they write.

[Skip Montreux]

That is a perfect analogy, Des.

[Des Morgan]

So this is not just an internal budgeting problem. It could certainly affect the companies selling the AI tools.

[Skip Montreux]

And that is where Wall street enters this story. Anthropic is headed for an IPO, and OpenAI is expected to follow. But to justify huge valuations to investors, they have to prove that their revenue can keep growing.

[Des Morgan]

And that revenue basically comes from selling access to AI computing power. The more tokens customers use, the more money these AI companies can make.

[Skip Montreux]

Exactly. OpenAI and Anthropic make money when customers use their AI system. And that usage is measured in tokens. So if major corporate clients like Microsoft, Amazon and Uber suddenly hit the brakes, cap their AI spending and focus only on useful output, that endless revenue growth hits a wall.

[Des Morgan]

So just as these AI companies are trying to go public and cash in, their biggest corporate clients are realizing the tech is too expensive.

[Skip Montreux]

Yes. And if the demand for AI was artificially pushed up just because employees were playing games with company money, the tech market is in for a very harsh reality check. If the token maxing party is over, these massive IPO ambitions and trillion dollar valuation hopes could be in serious trouble.

[Des Morgan]

And on that somewhat worrying note, I think it's time for us to get D2V down to vocabular.

[Skip Montreux]

The first item on our DTV list is the verb color code. When you color code something, you organize information by using different colors to show different categories.

[Des Morgan]

For example, you might color code the labels of your to do list. Red for urgent, green for finished, and yellow for the to do someday.

[Skip Montreux]

I used to color code my to do list, but I gave up everything. Ended up being colored red. That happens in the introduction of today's episode when Des was talking about his very expensive AI calendar app. He said the app reads all his emails, analyzes his tone, and then color codes his meetings based on their importance.

[Des Morgan]

In other words, the app uses different colors to show how important each meeting might be. Red meetings, very important, blue meetings, somewhat important, and green meetings, optional. To be honest, color coded meetings might be the most useful feature of the whole app.

[Skip Montreux]

I still think you are paying way too much for it. But the important thing here is that to color code means to use colors to organize or label information. Clearly.

[Des Morgan]

You might hear this in a project meeting. Someone might say, we've colour coded the supplier list so that green shows approved suppliers, yellow shows suppliers under review, and red shows suppliers with quality problems.

[Skip Montreux]

That is a very practical example.

[Des Morgan]

Moving on. Next on our D2V list are the phrasal verbs to burn through and to blow through.

[Skip Montreux]

Two very similar verbs, pretty much identical in meaning.

[Des Morgan]

They are they both mean to use something up very quickly, especially money, time or resources.

[Skip Montreux]

Right? And they usually suggest that something was used much faster than expected. You often hear these phrases with business nouns like to blow through a budget or burn through cash or burn through inventory.

[Des Morgan]

In today's episode, Skip first said that Uber burned through its entire annual AI budget in just four months. Then in the main report, I asked how exactly Uber managed to blow through its entire 12 month AI budget in just four months.

[Skip Montreux]

And again later in the report, I said Microsoft's developers were using agentic AI so much that the token cost blew through the company's AI budget in a matter of months.

[Des Morgan]

So in all those examples, the meaning is the same. The companies used up money that was supposed to last a full year much too quickly.

[Skip Montreux]

Exactly. The image is not just normal spending, it is fast spending. The money disappeared before the company expected it to.

[Des Morgan]

Kind of like me with a box of donuts.

[Skip Montreux]

That is probably the clearest example we've had all day. But in business, these phrases are often used when a company loses control of a budget or uses resources faster than planned.

[Des Morgan]

How might we use one of these phrases in a B2B context, skip?

[Skip Montreux]

Well, you might hear this in a meeting. Okay everyone, we burned through 80% of the total budget in the first six weeks. So we need to reduce the project scope before moving on to the next phase.

[Des Morgan]

Sounds like trouble. I wouldn't want to be the team leader of that project.

[Skip Montreux]

Our next DTV item is the idiom a canary in the coal mine. A canary in the coal mine is an early Warning sign that a bigger danger or problem may be coming.

[Des Morgan]

Unsurprisingly, this expression comes from coal mining. In the past, miners carried small birds called canaries into the mines with them to warn them if there was dangerous gas in the mine.

[Skip Montreux]

And how exactly did the canary warn them?

[Des Morgan]

Well, because canaries are more sensitive to toxic gas than humans, they would die sooner.

[Skip Montreux]

Wow, that is animal abuse.

[Des Morgan]

It was a different era in any case. Today, if we call something a canary in the coal mine, we mean it's the first sign of a larger problem.

[Skip Montreux]

And in today's episode, I used this expression. When we were talking about software developers and the huge cost of using agentic AI, I said coders were just the canary in the coal mine.

[Des Morgan]

In other words, coders are the first group impacted by the problem of overusing tokens. But the same issue could easily appear in other departments like Legal, Human Resources, Marketing and Finance.

[Skip Montreux]

The point is not that coders are the only employees facing the problem, but that they may be the early warning sign of a larger issue. How would you use this in another business context?

[Des Morgan]

Des, you might hear this in a sales strategy meeting. The sudden drop in repeat orders from our distributors may be a canary in the coal mine. It could be an early warning sign that demand for our product is weakening across the whole market.

[Skip Montreux]

That's a pretty strong example. It shows how the expression is used when one small, small problem may point to a much bigger problem coming later. What's our next word?

[Des Morgan]

The final item on our D2V list is the noun phrase reality check. A reality check is a moment when you are forced to stop, look at a situation honestly, especially after being too optimistic or unrealistic.

[Skip Montreux]

Right? It is often used when expectations are set too high and then new facts or new costs or poor results force people to rethink the situation.

[Des Morgan]

In today's episode, Skipt used this expression near the end of the report. He said that if the demand for AI was artificially pushed up because employees were playing games with the company money, then the tech market could be in for a very harsh reality check.

[Skip Montreux]

In other words, investors and AI companies may have believed that AI demand would keep growing forever. But if companies start limiting AI spending and focus only on useful output, the market may have to accept a harder truth.

[Des Morgan]

So the reality check is AI may still be powerful, but it may not be as cheap or as profitable as many people hoped. How might we use reality check in a different business context, Skip?

[Skip Montreux]

Hmm. Well, you might hear this in a management meeting. The failed pilot project was a reality check. For us, we realized our product was not ready for large scale use by our customers.

[Des Morgan]

Yeah, that's a good example. It shows that a reality check is not just bad news, it's a moment when a company has to face facts and make more realistic decisions.

[Narrator]

Would you like to help down to Business English reach more people wanting to improve their business English skills? Follow down to Business English on Apple Podcasts, YouTube, Music, Spotify, or any place podcasts are found. Leave a rating and a review and tell everyone how much you enjoy the show.

[Des Morgan]

Thanks for that report on Token Maxing Skip I'm still not totally sure I need to cancel my $120 AI calendar app to just yet, but I do have to admit the whole token maxing story has got me thinking about my AI usage.

[Skip Montreux]

It certainly should. What starts as a story about employees using too much AI quickly turns into a much bigger question about budgets, productivity, and whether companies are getting real value from these AI tools.

[Des Morgan]

I guess it comes down to whether AI is actually producing useful work D2B

[Skip Montreux]

members and Apple Podcast Subscribers the bonus down to Vocabulary for today's episode will be released within the next few days or so. In that bonus episode, we will look at six more useful words and phrases from today's report to pad something seamlessly behind the scenes, to hit the brakes, to jump on the bandwagon, and to

[Des Morgan]

cash in a very useful set of vocabulary for talking about AI spending, corporate budgets, and whether new technology is actually creating business value.

[Skip Montreux]

If you are a D2B member, make sure you have copied your Members Only podcast feed URL from your account page on the D2B website and pasted it into the podcast app of your choice. That way you will not miss out on the bonus DTV episode when it is released.

[Des Morgan]

And Apple Podcast subscribers, you do not need to do anything. The bonus DTV episode will appear automatically in your feed as soon as it goes live.

[Skip Montreux]

And if you are not yet a DTV member or Apple Podcast subscriber, but you get value from what we are doing here on down to Business English, please do consider supporting us by becoming a D2B member or Apple podcast subscriber.

[Des Morgan]

To become a D2B member, just visit d2benglish.com membership and sign up today.

[Skip Montreux]

That's D the number 2B English. And to become an Apple Podcast subscriber, just visit the down to Business English show page in Apple Podcasts and click the subscribe button. Thanks for listening everyone. See you next time.

[Des Morgan]

Bye bye.

[Narrator]

Down to Business English Business News to improve your business English.