Earlier this year, the message from tech companies to employees was clear: Use as much artificial intelligence in your work as possible.<\/p>\n
Employees called it \u201ctokenmaxxing<\/a>,\u201d with a token referring to a unit of A.I. use roughly equal to a word fragment. Companies like Meta and Amazon even encouraged workers to compete on leaderboards that tracked token use.<\/p>\n Then came the bills from companies, like Anthropic and OpenAI, that provide A.I. tools \u2014 and they were not cheap. Now the tokenmaxxing era appears to be over.<\/p>\n Meta told employees last week that it would soon limit A.I. use after seeing an \u201cexponential increase\u201d in costs. In May, Uber said it had blown through its projected A.I. spending for the year in just four months, and it has placed some monthly limits on A.I. coding tools. Walmart also set limits for different A.I. tools. And Amazon and Meta have taken down their tokenmaxxing leaderboards.<\/p>\n In other words, \u201ctokenminning,\u201d short for \u201ctoken minimizing,\u201d is now in.<\/p>\n The reversal, within just a few months, underlines how A.I. use remains in flux as people try to figure out how to best use the tools.<\/p>\n \u201cThe biggest problem is this is all changing so fast, people and companies don\u2019t know what to do,\u201d said Rob May, the chief executive of Neurometric, a start-up that helps companies better use A.I., and the author of \u201cThe Tokenminning Manifesto<\/a>.\u201d<\/p>\n \u201cC.E.O.s who did not know how to measure the A.I. savviness of their employees thought, \u2018Well, who\u2019s using the most tokens?\u2019\u201d he said, adding that the philosophy ended up promoting volume over efficiency.<\/p>\n OpenAI and Anthropic offer subscriptions that cost $10 to $200 a month for use of their A.I. models; when subscribers hit their usage limit, they are cut off. But the bulk of the revenue comes from offering tools to companies like Meta, Shopify and Amazon, which pay not only subscription fees but also for the tokens used by their tens of thousands of workers. So the more tokens that are used, the more money the A.I. costs.<\/p>\n A simple task, like asking A.I. to summarize the transcript from a company meeting, may use a few hundred tokens. More complex requests, like writing code to build a new product or feature, can use tens of thousands.<\/p>\n The costs of using A.I. models have soared as they have become more powerful and consume more tokens. Anthropic\u2019s newest A.I. model, Fable<\/a>, is twice as expensive as its previous model, Opus. While there are cheaper models, many employees have fallen into the habit of using the most powerful models for everything, Mr. May said.<\/p>\n The ways that people use A.I. have also changed. Instead of just conversing with A.I. chatbots, engineers deploy A.I. \u201cagents,\u201d which can work on complex tasks for hours at a time. As a result, engineers can use tens of thousands of dollars\u2019 worth of tokens each month.<\/p>\n Many companies said they were trying to be more strategic about A.I. spending after not seeing clear returns on their investment.<\/p>\n \u201cIf you\u2019re not actually able to draw a direct line to how many useful features and functionality you\u2019re shipping, that trade becomes harder to justify,\u201d Andrew Macdonald, Uber\u2019s chief operating officer, said in a recent podcast interview<\/a>. \u201cThat link is not there yet.\u201d<\/p>\n That\u2019s not to say companies won\u2019t keep spending big on A.I. Meta told employees that it was on track to spend billions on A.I. use this year, but wanted to \u201cfind places we can spend less while getting similar or better business results.\u201d Marc Benioff, the chief executive of Salesforce, the enterprise software company, said his company planned to spend hundreds of millions on A.I. this year but now tracked \u201cagentic work units\u201d instead of tokens. The new metric is supposed to measure output, not just use.<\/p>\n Meta\u2019s and Walmart\u2019s limits on employee A.I. use were reported earlier by The Information<\/a> and Bloomberg<\/a>.<\/p>\n It\u2019s unclear how \u201ctokenminning\u201d might affect the bottom lines of Anthropic and OpenAI. At the height of the tokenmaxxing era this year, the A.I. companies reported record revenues driven by the use of coding tools. Last week, Meta told its engineers to use its internal coding assistant, MetaCode, instead of third-party tools if possible.<\/p>\n