AI bots scale USDC crypto payments with new machine-payment rails
A Keyrock report says AI bots are moving from experimentation to scaled USDC crypto payments. From May 2025 to April 2026, AI agents are projected to complete 176M+ blockchain transactions worth over $73M.
Major firms are building machine-to-machine rails. Coinbase is pushing its x402 protocol, letting AI apps pay in USDC on-chain without accounts or subscriptions. Stripe targets the space with its Machine Payments Protocol (MPP) on Tempo. Google launched AP2 for authorizing AI spending, while Visa adds token-based authorization for AI-driven card purchases.
The economics favor USDC crypto payments. The report finds 76% of AI program payments are below the typical ~30-cent card fee, often in the 1–10 cent range. Stablecoin settlement on chains like Base or Tempo can drive fees to below $1 cent. It also notes concentration risk: 98.6% of machine-to-machine payments settle in USDC.
Regulation remains the key bottleneck. MiCA, the US GENIUS Act, and the EU AI Act are expected around mid-2026, but none directly addresses responsibility or identity for autonomous payments.
For traders, this supports USDC infrastructure momentum, but regulatory clarity and issuer concentration are likely to shape sentiment and risk appetite.
Neutral
Bullish use-case momentum is clear: scaled machine-to-machine payments in USDC, more payment rails from major fintech/crypto players, and very low per-transaction economics versus card rails. However, the report also flags material uncertainties—regulatory frameworks do not directly cover responsibility/identity for autonomous payments, and USDC settlement is highly concentrated (98.6%), which can increase perceived risk. Net-net, this looks more like infrastructure adoption support than an immediate, broad-based price catalyst for USDC.