Sequential and puzzle keys
Private keys 1, 2, 3… correspond to real Ethereum and Bitcoin addresses. The lowest ones have been swept countless times — but a famous puzzle keeps the high end interesting.
Scanning from key #1
Private key 0x0000…0001 is a valid key that controls a real address. So does #2, #3, and so on. These low keys have been discovered and emptied by researchers and bots many times over; anything sent to them is swept within seconds. The interesting question is how far up anyone has actually searched.
The Bitcoin Puzzle Transaction (2015)
In January 2015, an anonymous creator sent Bitcoin to 256 addresses whose private keys have deliberately increasing bit-length: puzzle #1 uses a 1-bit key, #2 a 2-bit key, up to #256. It was built as a public benchmark for how far key-search technology has progressed. In April 2023 the prizes were raised roughly tenfold; the outstanding bounty now totals on the order of ~950–990 BTC.
Keys up to about #70 have been solved, along with several higher ones. The active frontier is puzzle #71 and above. For puzzles where the public key is exposed, solvers use Pollard's kangaroo and baby-step giant-step algorithms to search a known, bounded range far faster than blind brute force.
A recurring twist: when a solver broadcasts their claim, the key is briefly visible in the mempool, and front-running bots have stolen prizes — as happened on puzzles #66 and #69.
Why full sequential scanning fails
Bounded ranges with exposed public keys are tractable. Scanning the entire 2256 space sequentially is not — you would exhaust the universe's energy budget long before reaching any funded modern wallet.