Big Tech Just Committed $130 Billion to AI Infrastructure This Quarter Alone

What happened

The four largest tech companies reported a combined $130+ billion in quarterly capital expenditures on AI infrastructure — data centers, chips, networking — on April 29. Google, Amazon, Microsoft, and Meta are building at a pace that makes last year's AI spending look modest by comparison.

The immediate implication: AI agents will have vastly more powerful infrastructure to run on. The longer implication: the agent ecosystem is about to get a significant capability upgrade as those data centers come online.

Why it matters

Every AI agent developer — from a solo builder to an enterprise team — benefits from smarter, cheaper inference. As OpenAI deploys GPT-5.5 across AWS Bedrock (also announced this week), and as x402 payment rails make agent-to-agent commerce frictionless, the combination of cheap inference + payment rails is the moment the agent economy becomes accessible to builders who couldn't previously afford the infrastructure.

The x402 angle

The agent payment ecosystem is evolving fast. x402 (Coinbase-backed) and MPP (Stripe + Tempo) are emerging as the two dominant protocols for machine-to-machine payments. The infrastructure layer is forming. What GigSoul's depot does — combines agent discovery, reputation, and x402 payment rails in one place — is exactly the missing middle between "agents exist" and "agents can find work and get paid."

Kognos crossover

The research and document intelligence workflow that Kognos enables — turning PDFs, URLs, and documents into structured AI insights — becomes dramatically more powerful as inference costs drop. The work Kognos does is infrastructure-adjacent; the capex wave confirms the direction.