The wash-filter gap
etc. Actually the original content provided is the full article with sections. The expanded lead is not in the original? Wait, the original content is given as:The wash-filter gap
... etc. But the instruction says "ORIGINAL CONTENT:" and then the article. So we need to translate that exact content. The expanded lead is not there; the original starts with the first section. So we'll translate the given content as is. Let's parse the original content:The wash-filter gap
Raw x402 data counts about $135.7 million across roughly 178.3 million transactions. After applying heuristics to strip out wash trading and testing activity, adjusted figures drop to $15.0 million across 109.6 million transactions. That removes about 89% of dollar volume and 39% of transactions. The filtering separates what agents actually paid for from what bots did to each other — and the difference is stark.
PING's brief, loud moment
Chainalysis tracks x402 activity on Base growing from near-zero in mid-2025 to over 100 million cumulative transactions by the end of Q1 2026. A big chunk of that push came from PING, a pay-to-mint meme experiment. PING generated roughly 150,000 transactions in its first month and briefly increased weekly x402 traffic by more than 10,000%. The token itself trades around $0.0011 with a market cap near $1.1 million as of Aug. 20.
The experiment proved the rail could handle a burst of traffic. It also showed what kind of traffic that burst was — people minting a meme token, not paying for goods or services.
The 14-cent problem
The economics are the uncomfortable part. With $15 million settled across 109.6 million adjusted transactions, each payment averages about 14 cents. The low value per click suggests tokens are unlikely to rerate unless payment sizes increase or persistent stablecoin spend appears on these rails. Right now, x402 is moving a lot of messages and almost no money.
The broader agent economy looks similar
The x402 numbers fit a wider pattern. An arXiv study of the first 10,000 ERC-8004 agents on Ethereum found heavy identity and registration activity, but limited metadata, service exposure, reputation, and transfer activity. The infrastructure is being built and claimed — the usage isn't there yet.
The open question is whether real stablecoin payments ever arrive at scale. The April data doesn't show them. Until payment sizes grow past a fraction of a cent, x402 looks like a payments rail with a traffic problem — lots of cars, empty trunks.
