A story we follow
Enterprise token-bill shock and spending controls in 2026
Follows the reported 2026 reckoning over enterprise AI token costs: CFOs seeing large bills in February and March, per-person caps, model routers and pullbacks from AI-first mandates, plus OutSystems' own spend peak and optimization. Does not cover general AI pricing or frontier-lab finances.
A story page follows one specific event across podcast discussions, with an overview and a timeline of what changed. It updates when new episodes discuss the event, and you can get those updates by email or push. How our formats work
By email or push, when we publish an update to this story. It does not subscribe you to other stories.
Overview
In 2026, companies that had pushed heavy AI use ran into large token bills and began reining in spending. Woodson Martin, CEO of OutSystems, said on an episode published October 7 that every CFO received a large token bill in February and March. In his account, companies responded with per-person caps, routers sending jobs to cheaper models, and in some cases pullbacks from AI-first-for-everything strategies. He said he thinks Microsoft, Meta and Uber made announcements of that kind, but he hedged and gave no details, and the claim rests on his recollection alone.
Martin described OutSystems' own spend peaking around June and July after a company-wide push to use models for everything, which he linked to faster feature releases. OutSystems then built a custom harness and an LLM router and, he says, is spending less than it forecast for Q3. He argues for platforms like his own as the way to control costs, and OutSystems sponsored that episode.
On Latent Space (published October 21), the host described a "token-maxing" era of token leaderboards followed by companies discovering their costs were too high, which he called entirely predictable. Tenex co-founder Alex Lieberman said Tenex worried about cost as headcount grew, but that his team never really bought into token-maxing. He placed peak concern six to eight months before the conversation and argued that falling per-task costs now make it much harder for heavy users to blow up a company's budget. A Tenex engineer separately estimated the firm's token spend at one to two times employee spend and credited Claude Opus 5.5 with recent sharp cost drops. These are the speakers' own accounts and estimates.
Recent coverage
Changes to our coverage published in the past seven days. These are site update dates; podcast discussion dates and sources are in each version.
No changes to this account were published in the past seven days.
What changed
Dates show when each podcast discussion was published.
-
New information
On Latent Space, the host described token leaderboards giving way to cost alarm. Tenex's Alex Lieberman said the firm worried as headcount grew, put peak concern six to eight months earlier, and argued that falling per-task costs now make runaway token spending much harder. A Tenex engineer estimated token spend at one to two times employee spend.
-
New information
Martin, speaking as OutSystems CEO, recalled that CFOs got shocking token bills in February and March 2026, prompting caps, routers and pullbacks from AI-first mandates, which he attributed with hedging to Microsoft, Meta and Uber. He said OutSystems' own spend peaked in June and July, with release velocity rising from four major features in Q4 to 26 in Q2, and then fell below its Q3 forecast after it built a custom harness and an LLM router. These are his own claims, and he argues for platforms like OutSystems as the way to control costs.
Podcast discussions
- Latent Space AI for the Fortune 2000: How Tenex Builds an AI-Native SDLC — Alex Lieberman & CJ HessOct 21, 2026
- The Cognitive Revolution Software That Never Breaks: OutSystems CEO Woodson Martin on Building Enterprise-Grade Apps at ...Oct 7, 2026
Sources
- 01
- 02
Our coverage
Tenex's Lieberman says runaway token costs are less of a worry now
Tenex estimates its AI token bill at one to two times its employee costs. Alex Lieberman says falling prices make runaway spending less of a worry than six to eight months ago, and make it harder for enterprises to justify fine-tuning.
OutSystems CEO says a harness and router cut its AI token bill
OutSystems CEO Woodson Martin says the company's AI token spending peaked in June and July, then fell below forecast after it built its own harness and model router. He argues most enterprise work doesn't need frontier models.
Version history
-
Oct 30, 2026 · Version 2
On Latent Space, the host described token leaderboards giving way to cost alarm. Tenex's Alex Lieberman said the firm worried as headcount grew, put peak concern six to eight months earlier, and argued that falling per-task costs now make runaway token spending much harder. A Tenex engineer estimated token spend at one to two times employee spend.