An AI agent running on Coinbase’s Base network analyzed SpaceX’s massive S-1 filing and produced a full investment committee memo in 12 minutes, spending just $1.87 in USDC across six API calls. The demonstration, led by Base Product Head Nick Prince, offers one of the clearest examples yet of autonomous AI agents using stablecoins as a native payment rail for real financial work.
What Happened: A Sub-$2 IPO Analysis on Base
Nick Prince, Senior Product Leader at Coinbase and head of Base product, ran the demo on agentic.market, a platform operating on Base where AI agents can pay for data and compute using on-chain micropayments. The agent ingested SpaceX’s 226 MB S-1 filing and made six x402 API calls to pull live market data, settling each payment in USDC on-chain.
The total cost came to $1.87 in USDC. For context, a Bloomberg terminal subscription runs roughly $24,000 per year.
On-Chain AI Analysis Cost
in USDC — total cost to run institutional-grade SpaceX IPO analysis
Completed in 12 minutes · Powered by AI Agents on Base
The output was not a simple summary. The agent produced a structured investment committee memo that included a fair equity valuation of approximately $345 billion, debt analysis flagging $42 billion in liabilities, a $6 billion-plus annual AI division burn rate, a $19.6 billion EchoStar spectrum commitment, underwriter conflict notes, bull and bear cases, and a final “HOLD” rating.
That $345 billion figure sits well below SpaceX’s reported IPO valuation target of $1.5 trillion to $2 trillion. The company is targeting a June 12 Nasdaq debut under the ticker SPCX, with Goldman Sachs, Morgan Stanley, Bank of America, Citigroup, and JPMorgan as underwriters. The AI memo flagged Morgan Stanley’s dual role, having also advised SpaceX on the xAI acquisition, as a potential conflict.
Circle CEO Jeremy Allaire responded publicly to the demonstration:
Incredible.
$1.87 in USDC paid by an agentic analyst to build a world class investment committee memo on SpaceX IPO.
The agentic economy has arrived. https://t.co/igRTr6V9ZC
— Jeremy Allaire – jerallaire.arc (@jerallaire) May 23, 2026
Source: @jerallaire on X
Why Stablecoins Are Becoming the Payment Rail for AI Agents
The demo’s significance extends beyond speed. The agent autonomously acquired and spent funds without a human intermediary managing API billing or credit card payments. USDC served as the settlement layer, with each of the six microtransactions clearing on Base’s low-fee infrastructure.
This model differs sharply from earlier AI-meets-crypto narratives that often lacked verifiable economic activity. The agentic.market platform has processed over $50.6 million in payment volume across more than 3.57 million transactions, with approximately 205,000 buyers and 39,000 sellers. That is real throughput, not a testnet exercise.
USDC’s programmability makes it suited for this kind of machine-to-machine commerce. With a market cap of roughly $76.29 billion and a stable peg at $0.9996, it provides the predictability agents need to budget and transact autonomously. MoonPay’s recent integration enabling crypto purchases directly within ChatGPT reflects the same convergence of AI platforms and crypto payment rails from a different angle.
Base’s fee structure is what makes the economics work. Sub-cent transaction costs mean an agent can make dozens of micropayments without the fees consuming the value of the work. The chain currently holds approximately $6.76 billion in total value locked, positioning it as one of the most active Layer 2 environments for this kind of application.
What Needs to Happen for Agent Finance to Scale
The SpaceX demo was a controlled showcase, not a production deployment. Moving from a single impressive run to a repeatable service requires several things: reliable data providers accepting x402 micropayments, agent frameworks with built-in wallet management, and trust mechanisms so users can verify an agent’s output quality before paying.
According to reporting from CoinGape, the x402 payment protocol allowed the agent to pay for API calls natively without API keys or human intervention, though independent technical verification of the protocol mechanics has not been published.
Competition is emerging. Other Layer 2 networks and agent platforms are exploring similar payment models, while AI agent frameworks like LangChain derivatives increasingly support on-chain integrations. The stablecoin layer is the common thread. Tether’s move to partner with the Georgian government on a new stablecoin shows how broadly the stablecoin infrastructure is expanding, even if the AI agent use case remains concentrated on USDC and Base for now.
The broader crypto market sits in cautious territory, with the Fear & Greed Index at 34, firmly in “Fear.” Yet the agentic economy narrative cuts against that sentiment. When an AI agent can replicate work that would take a human analyst days and cost thousands of dollars, the value proposition speaks for itself, regardless of whether Bitcoin is in a risk-on or risk-off cycle.
SpaceX’s IPO on June 12 will generate another wave of S-1 analysis. If agents like the one Prince demonstrated can commoditize that research in real time, the implications extend well beyond a single filing. The question shifts from whether AI agents can do institutional-grade financial work to how quickly the infrastructure, from regulated prediction markets to autonomous research platforms, catches up to the demand.
Disclaimer: This article is for informational purposes only and does not constitute financial or investment advice. Cryptocurrency and digital asset markets carry significant risk. Always do your own research before making decisions.
