a16z: The True Opportunity of Stablecoins is in Complementing, Not Disrupting

Original Title: Agentic commerce won't kill cards, but it will open a gap

Original Author: Noah Levine, a16z General Partner

Original Translator: Saoirse, Foresight News

A few weeks ago, an article published by Citrini Research claimed that stablecoins would bypass Visa and Mastercard, leading to a significant drop in the card networks' stock prices. The crypto community cheered.

The logic behind this is straightforward: AI agents will optimize each transaction, with fees being a kind of "tax," and stablecoins can bypass it.

I live and breathe crypto all day, and I wish this were true, but it is mostly wrong.

Not because stablecoins aren't significant, but because the real opportunity lies not in displacing cards but in serving merchants who struggle with traditional card acceptance.

Cards Will Capture the Majority of the Market

Citrini's argument hinges on one assumption: an AI agent free from human habits will actively optimize out card network fees.

But the credit card is more than just a payment tool. It offers unsecured credit, pre-authorizes uncertain transactions, and provides fraud protection through chargeback rights.

Stablecoins can facilitate transfers, but they fall short in these areas.

Imagine your AI agent books a hotel for you, and it looks nothing like the pictures.

With a credit card, you can dispute and get your money back.

With stablecoins, once the money is sent, it's gone.

82% of Americans hold reward credit cards (referring to credit cards with perks like cashback, points, air miles, hotel points, etc.), with a total of 18 billion cards in circulation worldwide.

For the vast majority of transactions, consumers are unlikely to willingly forgo purchase protection and rewards to opt for a payment method that offers no benefits and is irreversible.

Fraud detection is a significant advantage for card networks: these networks can run models in real-time on billions of transactions.

Stablecoins currently lack a network-level anti-fraud layer comparable to them.

Small payments are often cited as a weakness of credit cards, but card networks have long adapted to such mismatched transactions.

Visa has processed over 2 billion transit tickets by batching multiple swipes into daily settlements.

The card industry has never given up on any type of transaction; it will always invent new products to cover them.

There is also a question: "Smart agents can't hold cards."

But smart agents are essentially just new devices.

Your phone, watch, computer all hold independent tokens pointing to the same card, just like Apple Pay.

Phones have never gone through KYC; they just hold your token, and smart agents do the same.

Visa has issued over 16 billion tokens, and smart agents will also use these tokens.

Visa's Smart Business Framework is in a pilot phase, and Mastercard's Agent Pay has already launched for all U.S. cardholders.

The smart agent commerce protocol jointly developed by Stripe and OpenAI has integrated with Etsy, with over a million Shopify merchants set to go live.

The conclusion is clear:

For existing merchants and consumers, credit cards are almost destined to dominate smart agent commerce.

The opportunity for stablecoins lies elsewhere—in those merchants who have not yet appeared.

Those Yet to Appear Merchants

Every platform migration will always give rise to a wave of merchants that existing payment systems cannot serve.

When eBay emerged, individual sellers couldn’t open merchant accounts, so PayPal served them;

From 42,000 merchants in 13 years, Shopify grew to 5.5 million;

When Stripe was founded, many of its customers hadn't even been born.

The pattern remains consistent: winners serve the merchants that existing giants cannot underwrite.

The AI wave will generate these merchants faster than any previous platform migration.

Just last year, 36 million new developers joined GitHub.

In YC's Winter 2025 batch, a quarter of the companies had over 95% of their code written by AI.

On the popular AI coding platform Bolt.new, out of 5 million users, 67% are not developers at all.

People who couldn't write production-level code two years ago are now shipping software.

They are both buyers and sellers in the developer services market.

Imagine this:

A regular developer uses an AI tool to create a public company financial data dashboard in 4 hours. No website, no terms of service, no legal entity.

Another developer's bot hits it 40,000 times a week at $0.001 per call, generating $40 in revenue each week. No human ever sees a checkout page.

I see developers like this every week.

Their first question is always: How do I get paid?

For most, the answer is: You can't right now.

Existing payment infrastructure struggles to onboard these types of merchants.

It's not a technical problem; it's that payment companies have to underwrite the risk of these merchants once they start transacting.

If a merchant commits fraud or sees a high chargeback rate, the payment company is on the hook.

Tools with no website, no entity, no track record are almost impossible to get through risk review.

The system is working as designed; it's just not designed for this.

Payment companies can adapt, and they have in the past.

But it took PayPal 16 years from launch to the industry's first underwriting guidelines for payment service providers.

These new merchants need to get paid now.

For them, accepting stablecoins is like a street vendor taking cash only.

Not because cash is better, but because these merchants have historically struggled to get approved for card acceptance.

In this gap, stablecoins are the only viable option.

Despite rough wallets UX and still-forming compliance frameworks, protocols like x402 can embed stablecoin payments directly in HTTP requests:

No need for a merchant account, no need for a processor, no need for onboarding, no need to take on the risk of chargebacks.

These merchants aren’t choosing between stablecoins and credit cards.

They are choosing between stablecoins and not getting paid.

A New Commerce Is Emerging Here

Each new wave of merchants eventually gets absorbed by the traditional payment system; this time is likely no different.

But it is always: Merchants first, risk management follows.

During this interim period, stablecoins are the infrastructure.

· Card networks serve merchants that all payment acquirers can underwrite;

· Stablecoins serve merchants that no payment acquirer can underwrite.

The next wave of commerce emerges in this gap.

Original Article Link