Visa and Mastercard took a big 5%+ hit this week because a report by Citrini Research. My friend Simon Taylor lays out a summary of key points:

• Personal AI agents begin to re-shop insurance renewals and opt out of low-usage subscriptions automatically.
• This drives job cuts in those industries, which hits consumer spending, which hits the credit card sector.
• Professional services — law, tax prep, financial planning — see custom agent solutions displace those jobs.
• That further hits consumption. 75% of American GDP. Which hits consumer brand companies held by private equity.
• PE and private credit are massively exposed to SaaS and recurring revenue. Revenue that is no longer recurring in an AI world.
• The US economy is one long daisy chain of correlated bets on white-collar productivity growth. The top 20% of earners account for roughly 65% of all consumer spending. When those workers lose their jobs or take 50% pay cuts, the consumption hit is enormous relative to the number of jobs lost.
• The 2028 “future” scenario  note theorizes that agentic AI turned to stablecoin-based systems to execute transactions more cheaply
• Because AI agents optimize relentlessly for price and efficiency, they will bypass the 2-3% interchange fees of traditional credit cards by routing transactions through near-instant, near-zero-cost stablecoin rails on networks like Solana or Ethereum Layer-

My reaction? I just bought card networks stocks today. Why?

1. Citrini’s thesis fundamentally misunderstands the mechanics of global commerce, specifically the indispensable role of commercial agreements in defining how multilateral trust, risk, and value exchange occur. There is an entity behind every AGENT and the value the agent provides within a transaction and how it operates on behalf of WHO it represents is a commercial framework(s) in progress. I call this challenge the Gordian knot of monetization.
2. Multilateral means getting everyone’s buy in. Today Consumers, Banks and even most merchants prefer cards. How does an agent “save” 3% if the merchant doesn’t lower prices 3% and the consumer accepts no rewards? They all must agree to change something. While I’m a very big fan of x.402 and micropayment in stablecoin, NO ONE is predicting near term consumer or agent use in eCommerce (see Google UCP). PERHAPS an agent based tax accountant could asked to be paid in stablecoin, but that’s not in the Card networks current GDV.
3. Stablecoins are a Backend Transfer Mechanism, Not a Trust Architecture The Citrini report conflates the technological transfer of value with the commercial establishment of trust. Stablecoins are undoubtedly a powerful backend settlement innovation, allowing for near-instant, 24/7, peer-to-peer gross settlement. However, the technology of moving money is easy; creating certainty within commercial constructs with governance (measuring value and managing risk is hard). The economics of payments strongly incentivize the continued dominance of cards:
4. We are in a LONG stage of infancy for agentic commerce monetization. Just look at the BIG back steep of Google away from AP2 (mandates will be an internal protocol for next 2 yrs), to UCP and Embedded payments where the NRF merchants will own the checkout experience in an iFrame (15 yr old tech). 
5. Stablecoin Issuers prefer cards as the profitable off-ramp, earning interchange and allowing stablecoin balances to be spent anywhere cards are accepted without forcing the merchant to change their checkout infrastructure.
6. Citrini assumes a seamless transition to agent-to-agent economies and autonomous stablecoin wallets. In reality, the measurement of value in agentic commerce and the “Gordian knot” of monetization are complex structural problems that will take years to solve. The agent may seek to optimize price, but merchants don’t want price competition want to use their own agents (see yesterday’s WMT announcement on their agent driving 35% increase in basket size). 
7. Governance. While stablecoins can theoretically manage multi-merchant coordination and escrow, scaling this requires standardizing contracts, verifying identities, verifying authorization for action, and establishing automated dispute paths. These are challenges of governance, not just code. Ultimately, card networks are uniquely positioned to serve as the orchestration and permissioning layer for agentic commerce, providing the necessary rules and trust frameworks while utilizing stablecoins on the backend for efficient settlement. No one else is even close.
8. Remember when it was predicted that the internet would kill retailers? What happened? Have you seen Walmart’s stock?
9. Networks are hard to create and hard to destroy. My top message to the industry relating to this research is for Issuers. Stop with the friction in your successful networks. You aren’t going to build another one, the friction you create is where Stablecoin will attack first.
10. I feel confident in the 7 reasons I invest in Visa.