Meta will limit employee AI usage quota: self built gateway monitoring, internal costs may reach billions by 2026 According to Beating monitoring, Meta Platforms plans to control the soaring AI costs within the company by limiting employees' token usage quotas. According to a leaked internal memo, Meta is building a central gateway called AI Gateway to monitor employees' AI usage and expenses in real-time, set budgets, and set limits on token consumption. Meta expects that by 2026, the use of internal AI alone will generate billions of dollars in expenses. The quota measures are in stark contrast to Meta's previous strong promotion of AI. In November 2025, Meta notified employees that showcasing "AI driven impact" would be the core assessment criterion for 2026, and that performance bonuses would be linked to AI usage rates. Overpromotion has sparked a frenzy of "tokenmaxxing" among employees, with even an internal ranking called "Claudeomics" appearing to publicly display usage rankings. Before the ranking list was shut down, the total amount of tokens consumed by employees within 30 days skyrocketed to 73.7 trillion. Meta's Chief Technology Officer Andrew Bosworth subsequently issued a warning, emphasizing that simply increasing token consumption does not necessarily mean output, and employees should use AI tools in scenarios that can effectively improve efficiency. In order to further reduce expenses, Meta has begun to shift its internal AI development focus towards self-developed tools. The leaked memo shows that Meta is pushing its employees to gradually stop using third-party programming tools such as Claude under Anthropic and instead use its self-developed programming assistant MetaCode (formerly known as Devmate). The newly established application AI engineering department of Meta has been requested to fully improve MetaCode by generating programming challenges to produce high-intensity reinforcement learning training data. Although Meta still allows employees to access external models, stricter budget and quota approval mechanisms will be implemented in the self built gateway in the future. Meta is not the only company facing financial pressure due to excessive use of large models. In early 2026, companies such as Uber and ServiceNow exhausted their annual Anthropic quota in just a few months. ServiceNow has implemented daily usage monitoring for employees, and some venture capital firms have also started setting daily consumption limits for internal AI Token usage to prevent uncontrolled expansion of computing power costs. [Original link]