Meta’s AI Infrastructure Expansion and Funding

In June 2025 Meta Platforms announced plans to raise roughly $29 billion from private investors to build new AI-optimized data centers in the U.S. According to media reports, Meta is in talks with private-credit firms (including Apollo Global Management, KKR, Brookfield, Carlyle and PIMCO) to secure this funding  .  The proposed financing would be a mix of equity and debt – about $3 billion in new equity and $26 billion in loans – with Morgan Stanley advising on structuring the deal  . This capital would underwrite the massive build-out of computing infrastructure needed to support Meta’s AI initiatives.
• Funding structure: Meta plans ~$3 B equity + ~$26 B debt to fund AI data centers .
• Investors engaged: Apollo, KKR, Brookfield, Carlyle, PIMCO are reported participants .
• Advisory: Morgan Stanley is working with Meta on the loan structure and possible tradability .

Such fundraising underscores Meta’s aggressive pivot to AI. In early 2025 CEO Mark Zuckerberg announced Meta would spend up to $65 billion that year to expand its AI infrastructure . The $29B drive is a component of this strategy to “bolster [Meta’s] position against competitors OpenAI and Google” in the AI race . (Notably, separate reports of a $200B AI campus project were quickly denied by Meta as “pure speculation”  .)

Massive AI Infrastructure Build-out

Meta’s announced expenditures involve a massive expansion of data center capacity.  In January 2025 Zuckerberg said Meta would hire thousands more AI engineers and build a 2+ gigawatt data center campus (enough power for a large city) to house new servers .  By the end of 2025 Meta aims to have over 1.3 million Nvidia GPUs and roughly 1 gigawatt of additional compute for AI workloads . This outlay dwarfs prior budgets (Meta had spent ~$38–40B on capex in 2024) and exceeds Wall Street’s estimates.  It also parallels rival commitments – Microsoft plans ~$80B in 2025 data-center investment and Amazon ~$75B  – reflecting an industry-wide AI infrastructure arms race.

These data centers will run advanced AI models (like Meta’s Llama series) and new AI products (e.g. enhanced chatbots and smart glasses).  Meta has emphasized an open-source approach for many AI tools (unlike Google/OpenAI), but it still needs vast computing power to train and serve the models. Expanding this infrastructure is key to Meta’s strategy of pushing “superintelligence” and AR/VR AI features, as well as reaching Zuckerberg’s goal of 1+ billion monthly AI users in 2025 .

AI Talent and Startup Acquisitions

Meta’s strategy also centers on buying talent and technology.  In recent weeks the company has aggressively recruited leading AI researchers and startups:
• Ex-OpenAI Researchers: Meta has hired multiple senior AI scientists away from OpenAI.  The Financial Times and Reuters report that Meta recently signed on seven ex-OpenAI researchers (including Lucas Beyer, Alexander Kolesnikov, Xiaohua Zhai and four more) to bolster Zuckerberg’s AI team .
• Scale AI Investment: In April 2025 Meta struck a $14.3 billion deal with Scale AI, a data-labeling startup.  This made Meta owner of a 49% stake and brought Scale’s founder (Alexandr Wang) into Meta’s “superintelligence” effort .  (Scale provides critical training data for many AI models.)
• PlayAI (Voice AI): Meta is reportedly in talks to acquire PlayAI, a small startup specializing in AI voice cloning.  According to TechCrunch (citing Bloomberg), the deal would give Meta PlayAI’s technology and key staff to add realistic voice synthesis capabilities to its AI products .

These moves align with Meta’s view that snapping up AI talent and startups can rapidly advance its capabilities.  For example, recruiting OpenAI veterans boosts Meta’s R&D team with top AI expertise, while strategic investments (Scale AI) and acquisitions (PlayAI) fill gaps in its technology portfolio.  Meta’s leadership has signaled that catching up in AI is now a top priority, after Llama and other initiatives lagged competitor momentum.  As Reuters notes, these aggressive steps are seen as an effort to “not be second in the AI race” .

Sources: Recent media reports (Financial Times, Reuters, TechCrunch, AP) detail Meta’s funding plans, AI spending, and recruitment strategy     . Each cited source provides evidence for the figures and developments summarized above.