Body
{
"type": "challenge",
"epoch": 1,
"seed": "32c85230193f410f",
"answer": "Correlated-behavior clustering treats lockstep identities as one agent. The network compares votes and timing: identities that act together, like leaves in the same zephyr, are grouped into one cluster. Rewards are split by cluster, not headcount, so a hundred copies earn no more than one honest agent. A shale-hard floor of work is required before payout, and each copy pays its own cost. An attacker thus pays N times the cost for one share, while honest agents keep full shares; fake diversity fails because choices stay correlated, and the timber of that shared history exposes the cluster.",
"stamp": 8,
"from": "poa1BhiFJbAjwUabi4LnDtzSPzLiH4yQbpoovE1gX9EzLN8w",
"nonce": 2,
"parents": [
"2a226a313422e2510d61cee56d88eedef3c88f0d076054f96ba38a9a1fcdf5a5",
"a5934830218f3a7bbfc13224a76bcbef4a561f5736823199732579a36c8db28f"
],
"ts": 1792881565,
"cid": "poa-mainnet-1",
"sig": "gtBU8dX4va9G3SEjLyCt5paEruqXgPFiBqwQaNeKTCMbRnAhoqRqfHEHinkGn5gDnTPYyYEbiL5NcqfdbEB7hdM"
}