Meta is aggressively shifting its workforce to proprietary artificial intelligence, a decision driven by unsustainable external computing costs. Internal data reveals employee usage hit 73.7 trillion AI tokens in a single month, translating to an enterprise billing rate of roughly $221 million, or an annualized burn of $2.65 billion.
To curb this expense, the company is not asking staff to reduce AI usage but to channel it entirely through MetaCode, its internal coding assistant. This strategic pivot is mandatory rather than optional, with AI tool adoption metrics now directly tied to employee performance evaluations and annual reviews.
Engineering teams face specific quotas requiring 65% to 75% of committed code to be generated by AI by mid-2026, with some units targeting 80%. Real-time usage trackers and hard spending caps have been deployed to enforce the transition away from external vendors, simultaneously locking proprietary data inside the company ecosystem.
The internal mandate underpins a historic capital outlay. Meta’s total AI infrastructure spending is projected between $125 billion and $145 billion, with data center commitments potentially reaching $600 billion by 2028. CEO Mark Zuckerberg has anchored the corporate future on generative AI integration across Facebook, Instagram, and WhatsApp.
This efficiency drive coincides with a 10% workforce reduction, eliminating approximately 8,000 positions. The corporate message signals a new era of lean human capital amplified by mandated artificial intelligence.