Meta introduces AI Gateway with budget controls and token allocations for AI model usage, illustrating corporate innovation i

Editorial illustration for Meta to tighten AI token use with budgets, allocations and new AI Gateway

Meta to tighten AI token use with budgets, allocations...

Meta to tighten AI token use with budgets, allocations and new AI Gateway

By AI Daily Post Edited by Brian Petersen, Editor-in-Chief

June 13, 2026 • 3 min read

Meta is tightening the reins on its internal AI spend after an internal memo warned that usage is soaring. Sent to roughly 6,000 staff, the note flagged an “exponential increase” in token consumption and warned the company is on track for billions of dollars in AI costs by 2026—just from internal projects. Employees and teams have had little visibility into how many tokens they’re burning, let alone control over the budget.

Starting in 2027, Meta will roll out a token‑management system that includes budgets, allocations and a new “AI Gateway” dashboard built by a dedicated engineering team. The tool will consolidate usage data, surface spending in one place and soon fire automatic alerts for abnormal spikes. At the same time, the firm is nudging staff away from third‑party models like Anthropic’s Claude toward its own coding assistant, MetaCode, even though Meta’s models aren’t yet front‑running the field.

The shift follows a period of “tokenmaxxing,” where a leaderboard called “Claudeonomics” spurred employees to log 73.7 trillion tokens in just over a month, a practice that CTO Andrew Bosworth has now publicly discouraged.

Starting in 2027, Meta plans to manage AI tokens more tightly with budgets, allocations, and dedicated tools. A team of developers and engineers built a central dashboard called "AI Gateway" that tracks usage and spending in one place. Automatic alerts for unusual cost spikes are coming next.

Meta also wants to steer employees away from third-party tools like Anthropic's Claude and toward its own coding assistant, MetaCode. Other models will still be available, though; Meta's own models aren't yet competitive at the frontier. Token usage doesn't equal productivity Engineers in Meta's new "Applied AI Engineering" division are working to improve MetaCode by creating coding tasks as training data.

Earlier, Meta had made AI usage a "core expectation" in performance reviews, which led to so-called "tokenmaxxing": employees artificially inflated their consumption through an internal leaderboard called "Claudeonomics," racking up 73.7 trillion tokens in just over 30 days. CTO Andrew Bosworth pushed back in a separate memo: "Nobody should be using AI tools just for the sake of using them. All motion is not progress and token usage alone is not a measure of impact of any kind." Tools should be used when they "genuinely allow us to do better work, faster." Amazon ran into a similar tokenmaxxing problem that spiraled out of control.

That both companies are now reining in AI spending fits a broader pattern: businesses are questioning whether AI is actually boosting productivity.

Meta shifts from "tokenmaxxing" to token managing as internal AI costs reportedly hit billions - THE DECODER

Why this matters

Meta’s new AI token controls signal a shift from unchecked experimentation to cost‑centered governance. An internal memo warned that internal AI usage is climbing exponentially, projecting billions in spend by 2026. Can tighter budgets coexist with the rapid iteration that many of us rely on?

Starting in 2027 the company will enforce budgets, allocations and a central “AI Gateway” dashboard that consolidates usage data and triggers alerts for abnormal spikes. For developers, this means we may soon need to request token allowances before running large models, adding a layer of administrative overhead. Founders could see tighter spend visibility, but the effectiveness of automatic alerts is still unproven.

Researchers might find the centralized view useful, yet it could also constrain rapid prototyping if limits are applied too rigidly. We appreciate the attempt to curb runaway costs, but it remains unclear whether the new tooling will balance fiscal discipline with the flexibility that many AI projects require. Our community will have to watch how Meta’s internal controls translate into external developer experiences.