--- title: Token costs are doubling every 45 days and productivity... description: AI token costs at leading companies are doubling roughly every 45 days, while productivity gains hover around 5%. The most advanced AI-adopting companies were... canonical: https://daily.dev/posts/token-costs-are-doubling-every-45-days-and-productivity-is-up-5--gwa82lqjn twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Token costs are doubling every 45 days and productivity is up 5% | daily.dev og:description: AI token costs at leading companies are doubling roughly every 45 days, while productivity gains hover around 5%. The most advanced AI-adopting companies were... og:url: https://daily.dev/posts/token-costs-are-doubling-every-45-days-and-productivity-is-up-5--gwa82lqjn og:image: https://api.daily.dev/og/posts/gwa82lqJn.png og:image:alt: Token costs are doubling every 45 days and productivity is up 5% og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Token costs are doubling every 45 days and productivity is up 5% **[Trends](https://daily.dev/sources/trends)** · 3 min read · 2 upvotes · 0 comments ## Summary AI token costs at leading companies are doubling roughly every 45 days, while productivity gains hover around 5%. The most advanced AI-adopting companies were already spending ~$90K per employee annually on AI as of May 2026, a figure that could exceed the average software engineer's $192K salary before year-end. Meta shut down an internal token spend leaderboard after costs approached billions, Uber burned its entire AI coding budget by April, and Microsoft cancelled Claude Code licenses. The piece explores the tension between optimistic narratives (AI enables growth and hiring) and skeptical ones (costs compound faster than gains), and raises a novel organizational risk: engineers gaming agentic loops to appear productive without delivering value. Token caps per engineer, proportional to demonstrated ROI, are predicted to become standard budget practice. ## Content The numbers are getting hard to ignore. Meta hit 60 trillion tokens in a single month. Uber burned through its entire annual AI coding budget by April. The most aggressive AI adopters are already spending around $90,000 per year on AI per company, and at 14.1% monthly growth, that figure could exceed the average software engineer's $192K salary before 2027. Adam Mosseri put it plainly: within a year or two, companies will need to cap token spend per engineer the way they cap payroll. Meta already killed an internal token leaderboard when costs approached billions. Microsoft cancelled Claude Code licenses. The era of "just let people use it" is ending. What's making this hard is that the costs are almost completely invisible. Tailscale's Aperture gateway recently started letting developers see what their $20/month Claude Code subscription actually costs per request. A simple greeting: $0.31. A browser game: $3.29. A complex home automation project on Opus: $32.76. All of it hidden under a flat fee, all of it adding up. The Ramp CEO's take is blunt and probably right: just showing people their spend makes them more efficient. "Exposing people and saying, 'Here's your AI spend,' they start getting more efficient." Nobody's trying to waste money; they just can't see it. There's a darker read too. One developer on Twitter noted that AI costs are doubling every 45 days while productivity gains are running around 5%, then added: "there's a unique kind of employee sabotage when you're paid just the same, scared about layoffs, and expected to do 10x more work with AI. So you start running agentic loops as the ultimate token burn trick." The engineering response is already taking shape. Monday.com published a breakdown of three stacked compression tools that cut output tokens by 65-95%. Ratel, an open-source context layer, claims ~80% token reduction by only loading relevant tool schemas per turn rather than dumping everything into context upfront. One developer rebuilt their entire research pipeline after burning a full Claude team plan in 30 minutes, finding that harness choice alone can cause a 66x token spread for the same model. Qualcomm's CEO is out here saying agentic AI will require "gazillions" of tokens, which is technically true and completely unhelpful for anyone trying to set a budget today. The real shift is simpler: token spend is becoming a line item that CFOs will own, and the developers who understand where the waste actually lives are about to become very useful people to know. ## Similar posts on daily.dev - [Sticker shock has execs rethinking this whole AI thing](https://daily.dev/posts/sticker-shock-has-execs-rethinking-this-whole-ai-thing-ywgulwvsx) · The Register · 0 upvotes · 0 comments - [The token bill comes due: Inside the industry scramble to manage AI’s runaway costs](https://daily.dev/posts/the-token-bill-comes-due-inside-the-industry-scramble-to-manage-ai-s-runaway-costs-glxojg09s) · TechCrunch · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#productivity](https://daily.dev/tags/productivity), [#ai-agents](https://daily.dev/tags/ai-agents), [#ai-coding](https://daily.dev/tags/ai-coding) [View this post on daily.dev](https://daily.dev/posts/token-costs-are-doubling-every-45-days-and-productivity-is-up-5--gwa82lqjn) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Token costs are doubling every 45 days and productivity is up 5%","url":"https://daily.dev/posts/token-costs-are-doubling-every-45-days-and-productivity-is-up-5--gwa82lqjn","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/token-costs-are-doubling-every-45-days-and-productivity-is-up-5--gwa82lqjn"},"datePublished":"2026-07-15T14:56:48.273Z","dateModified":"2026-07-22T14:01:56.760Z","description":"AI token costs at leading companies are doubling roughly every 45 days, while productivity gains hover around 5%. The most advanced AI-adopting companies were...","isAccessibleForFree":true,"articleSection":"Trends","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Trends","logo":"https://media.daily.dev/image/upload/s--ZfSp3asX--/f_auto,q_auto/v1780996004/logos/trends?_a=BAMAMiWQ0","url":"https://daily.dev/sources/trends"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/token-costs-are-doubling-every-45-days-and-productivity-is-up-5--gwa82lqjn","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,productivity,ai-agents,ai-coding","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Trends","item":"https://daily.dev/sources/trends"},{"@type":"ListItem","position":3,"name":"Token costs are doubling every 45 days and productivity is up 5%"}]} ```