May 29May 29 An AI consultant told Axios that one of their clients "recently spent half a billion dollars in a single month" on Claude, the Anthropic AI assistant, because nobody had capped how much its employees could run on the licenses they were given
May 29May 29 Author https://boingboing.net/2026/05/29/a-company-accidentally-spent-500-million-on-claude-in-one-month.html/amp
May 29May 29 5 minutes ago, kevinUsa said:An AI consultant told Axios that one of their clients "recently spent half a billion dollars in a single month" on Claude, the Anthropic AI assistant, because nobody had capped how much its employees could run on the licenses they were givenBro I sent 50$ for API rather than scubscription, email chese robo reply ichindi one of human will address ani. Is there a phone number to talk about this to Claude?
May 29May 29 7 minutes ago, kevinUsa said:An AI consultant told Axios that one of their clients "recently spent half a billion dollars in a single month" on Claude, the Anthropic AI assistant, because nobody had capped how much its employees could run on the licenses they were giveninka enni layoffs chesi cover cheyali ah bokka ni
May 29May 29 msft ani talkchala projects ki ee year funding release avale, although end of MayMSFT fiscal year starts July 1st, by now projects ki budgets reveal avutaai, kani inka avaleduAndaru Employees ni contractors ni Claude vadoddu ani mail kuda vachhindi anta
May 29May 29 Just now, pichhipullayya said:msft ani talkchala projects ki ee year funding release avale, although end of MayMSFT fiscal year starts July 1st, by now projects ki budgets reveal avutaai, kani inka avaleduAndaru Employees ni contractors ni Claude vadoddu ani mail kuda vachhindi antaIt’s Amazon.
May 29May 29 Author An AI consultant told Axios that one of their clients "recently spent half a billion dollars in a single month" on Claude, the Anthropic AI assistant, because nobody had capped how much its employees could run on the licenses they were given.Corporate leaders "are starting to question whether soaring AI spending is delivering meaningful returns." One reason the bill got so big: agentic AI tools, the kind that run multi-step tasks on their own, "eat up 1000x more tokens than querying an LLM."Tom's Hardware notes that some of that spending goes to automating "dreary and mundane tasks they don't like doing, rather than valuable or meaningful work," with some employees "even using AI models to check the weather."X users guessed the mystery client might be Amazon, where staff were reportedly caught inflating their AI token use to hit internal targets — a practice nicknamed "tokenmaxxing." The Financial Times reported that Amazon scrapped an internal AI usage leaderboard after employees started doing needless tasks to climb it. Uber's CEO said there was no link between tokenmaxxing and shipping anything useful.The company remains unnamed. As Tom's Hardware puts it, "the very scale of the overspend narrows it down to only the very largest corporations globally.
May 30May 30 4 hours ago, pichhipullayya said:msft ani talkchala projects ki ee year funding release avale, although end of MayMSFT fiscal year starts July 1st, by now projects ki budgets reveal avutaai, kani inka avaleduAndaru Employees ni contractors ni Claude vadoddu ani mail kuda vachhindi antathey own open ai and using rival product? its same like their employees using mac at workplace
May 30May 30 4 hours ago, pichhipullayya said:msft ani talkchala projects ki ee year funding release avale, although end of MayMSFT fiscal year starts July 1st, by now projects ki budgets reveal avutaai, kani inka avaleduAndaru Employees ni contractors ni Claude vadoddu ani mail kuda vachhindi antaMsft lo pani chesey andaru copilot ni galiki vadilesaru. Ippudu emo claude vaadi bill ki pee pee chesthunnaru
May 30May 30 Author The $500 million number is almost certainly real but may not mean what it sounds like. Anthropic’s published API pricing for Claude Opus puts output tokens at $75 per million. To reach $500 million in a single month at that rate you would need roughly 6.67 trillion output tokens, which is an enormous number that would require either an extraordinarily large workforce of heavy users, an automated pipeline running continuously at massive scale, or both.
May 30May 30 Author Firms like Amazon, Uber, and Microsoft are pulling back on generative Al after unchecked spending led to massive overruns. Amazon axed its 'Kirorank' leaderboard that encouraged wasteful Al tasks, Uber exhausted its 2026 Al budget months early, and Microsoft canceled Claude licenses in favor of its own tools. Executives now demand proof of productivity gains before more spending, marking a shift from hype-driven adoption totargeted efficiency.
May 30May 30 5 hours ago, Gorantlamdhav said:they own open ai and using rival product? its same like their employees using mac at workplaceMSFT also has a stake in Anthropic.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.