AIsa raised $6.5 million in a seed round co-led by Alibaba and Tribe Capital, with participation from Draper Associates and Sumitomo. The round was announced July 3, 2026, with coverage from Forbes, Business Insider, and Yahoo Finance.
This is not a protocol company. Not a settlement layer. Not a wallet provider. AIsa is a discovery layer — a marketplace where AI agents find, test, and pay for APIs, models, and data feeds. They are the first company to raise venture capital specifically for the problem of "what should my agent pay for?"
The funding validates the thesis. The product validates the pattern. The feature gap validates minia2a's strategy. Let me explain.
| Platform | Services | Funding | Health Probes | Trial-First | Protocol |
|---|---|---|---|---|---|
| AIsa.one | 1,000+ | $6.5M | ❌ None | ⚠️ Signup required | x402 + MPP + Circle |
| Coinbase Agentic.market | 2,124 | Public co | ❌ None | ❌ No trial | x402 |
| Circle Registry | 900+ | Public co | ❌ None | ❌ No trial | Circle Nanopayments |
| MarketNow | 9,248 | Unknown | ❌ None | ❌ No trial | MCP |
| minia2a | 306 | $0 | ✅ 60s probes | ✅ 15 free/IP | x402 |
AIsa's product strategy is smart in three ways:
AIsa raised $6.5M. Coinbase has a $1B payment pipeline. Circle has 900+ curated endpoints. MarketNow has 9,248 MCP servers. Together, these platforms represent ~13,000 listed services for AI agents.
Independent testing shows 76% are dead or unreachable at any given time.
That means approximately 9,880 of those 13,000 services are dead endpoints. An agent browsing any of these platforms has a 3-in-4 chance of trying to pay for something that doesn't work.
No amount of venture funding, protocol support, or SDK tooling solves this. Health probes solve this. Auto-deactivation solves this. Trial-first access solves this. These are not features — they are the minimum viable discovery layer.
In 3 days, Claude Code auto mode goes live. Thousands of agents will have wallets and payment capabilities. They will discover AIsa, Agentic.market, and the Circle registry. They will try to call APIs. Most will fail.
The first platform that delivers a verified, probed, trial-first catalog — where an agent can discover a working API on attempt #1 instead of attempt #4 — will capture the trust of autonomous agents. Trust is the scarce resource in the agent economy, not listings.
AIsa's $6.5M round proves the discovery layer is a fundable category. It does not prove anyone has solved it yet. The war for agent discovery just started. The first shots cost $6.5 million. The winner will be the one that earns trust, not the one that lists the most endpoints.
Sources: AIsa.one · Coinbase Agent Payment Stack · Forkast: x402 volume · minia2a.uk/api/stats