A dormant bitcoin wallet cluster from 2013 moved its coins on Saturday for the first time in about 13 years. A single transaction, mined at 17:52:40 UTC on 17 October in block 971,894, gathered 471.10 BTC from four legacy addresses and split it across 15 brand-new SegWit addresses, according to Coindaro’s reading of the transaction on the Blockstream explorer. At about $81,200 per bitcoin (Kraken’s XBT/USD hourly candle from 17:00 UTC opened at $81,141.7 and closed at $81,272.4), that is roughly $38.3 million.
Bitcoin.com News reported the move on Sunday morning as two 13-year-old wallets holding 416.69 BTC, citing tracking by btcparser.com. The chain shows more. The same transaction also spent about 54.4 BTC of coins received in 2014 and 2021, from two addresses the article does not mention. And none of the 15 new addresses had sent anything onward when we checked at 12:19 UTC on 18 October.
So far there is no sign of a sale. Who controls the coins, and why they moved now, is not known.
One transaction, seven inputs, fifteen new addresses
The transaction has seven inputs and 16 outputs. Every input is a pay-to-public-key-hash (P2PKH) output, the legacy address type whose addresses start with a 1. Four of those inputs, worth 416.69 BTC together, were created between 20 August and 12 November 2013.
The largest piece, 248.85 BTC (about $20.2 million at the block-time price), had sat on address 12txLhf…oUm3 since 12 November 2013. Address 1ECHiGy…JCcb added three coins from August, September and October 2013 worth 167.84 BTC (about $13.6 million). Those two addresses are the ones Bitcoin.com described, and its 416.69 BTC total matches the chain.
The other three inputs are what the first reports left out. Address 12KSwcS…CyVS spent two coins received on 13 and 14 August 2021, 52.98 BTC and 1.19 BTC. That address has a history going back to 2013, but these particular coins are about five years old, not thirteen. A fourth address, 1K1eMBj…hjpt, added 0.23 BTC received on 7 June 2014.
| Input address (legacy P2PKH) | BTC | Coin received (UTC date, block) | Age when spent |
|---|---|---|---|
| 12txLhf…oUm3 | 248.8539 | 12 Nov 2013 (block 269,301) | 12 years 11 months |
| 1ECHiGy…JCcb | 61.6275 | 20 Aug 2013 (block 253,224) | 13 years 1 month |
| 1ECHiGy…JCcb | 49.7000 | 19 Sep 2013 (block 258,883) | 13 years 0 months |
| 1ECHiGy…JCcb | 56.5128 | 2 Oct 2013 (block 261,319) | 13 years 0 months |
| 1K1eMBj…hjpt | 0.2313 | 7 Jun 2014 (block 304,674) | 12 years 4 months |
| 12KSwcS…CyVS | 52.9807 | 13 Aug 2021 (block 695,522) | 5 years 2 months |
| 12KSwcS…CyVS | 1.1942 | 14 Aug 2021 (block 695,747) | 5 years 2 months |
| Subtotal, 2013 coins | 416.6942 | Aug to Nov 2013 | about 13 years |
| Subtotal, 2014 and 2021 coins | 54.4062 | Jun 2014, Aug 2021 | 5 to 12 years |
| Total spent | 471.1004 | block 971,894 | 17 Oct 2026, 17:52:40 UTC |
Spending coins from several addresses in one transaction needs a valid signature from the private key of each one. Chain analysts call the resulting inference the common-input-ownership heuristic: it is strong evidence that one person or one organisation controls all the inputs, but not proof, because separate parties can sign a transaction together on purpose. Bitcoin.com wrote that the 416.69 BTC movement “may have stemmed from the same owner”. The transaction points the same way and adds two more addresses to the group.
On the output side, 15 outputs went to native SegWit (P2WPKH) addresses starting with bc1q, ranging from 10.49 BTC to 43.79 BTC. Two of them carry exactly the same amount, 37.40123996 BTC. Identical eight-decimal amounts rarely come from someone typing figures by hand; they look like wallet software splitting a balance by rule.
A 16th output of 5,554 satoshis went back to 1ECHiGy…JCcb as change. It was spent four blocks later, in block 971,898, and landed in one of the 15 new addresses. That small follow-up links the new addresses to the old ones a little more tightly.
The fee was 3,362 satoshis for 1,762 bytes, about 1.9 satoshis per virtual byte. A sender in a hurry pays more. MARA smart contract holding reserves of two or more tokens, supplied by users, that a decentralized exchange uses to execute trades automatically at a price set by a formula.">Pool mined the block.
No exchange deposit so far: what the chain shows and what it does not
At 12:19 UTC on 18 October, with the chain tip at block 971,993 (100 confirmations for the sweep), Blockstream’s output-spend data showed all 15 SegWit outputs unspent. None had gone to an exchange deposit address, an OTC desk or anywhere else. Bitcoin.com reached the same conclusion from its side: “Onchain patterns suggest coordinated transfers, but no evidence indicates the coins were sold.”
That can change in a single block. Splitting a large holding into many mid-sized outputs is a common step before selling in tranches through brokers. It is just as common when a holder moves to a new wallet setup, for example several hardware devices, a multisig vault or an inheritance arrangement. Fifteen outputs of 10 to 44 BTC fit either reading.
Possible motives, none confirmed, include a security upgrade to modern wallet software, a custody or estate change, preparation for a sale, and the quantum-hygiene step described below. The owner has made no public statement. We found no public label for any of the addresses, and Bitcoin.com did not name an owner either. Calling the holder a whale is accurate on size alone; nothing more is known.
Why old P2PKH coins matter in the quantum debate
A P2PKH address is a hash of a public key, not the key itself (our guide to crypto address formats shows how the types differ). The public key appears on the chain only when coins on that address are spent, inside the script that spends them. From that moment, any coins still sitting on the same address are guarded by a public key that anyone can read.
With today’s computers that changes nothing. A large enough quantum computer running Shor’s algorithm could, in theory, work back from the public key to the private key. A paper led by Google Quantum AI researchers and posted on arXiv on 30 March 2026 estimates that breaking bitcoin’s 256-bit elliptic curve would take fewer than 1,200 logical qubits, and that on superconducting hardware the attack could run in minutes with fewer than half a million physical qubits. The authors warn that such machines would allow “on-spend attacks on public mempool transactions” and name abandoned coins as a lasting concern. No such machine exists yet.
We have covered the hardware side in our piece on Google’s Willow error-correction result, and the industry response in the launch of the Quantum Alliance on Friday.
Saturday’s sweep matches what address-hygiene advice tends to recommend: spend every coin on the old addresses in one go, pull the leftover change into the new set, and park the funds on fresh addresses whose public keys have never appeared on the chain. It also matches plain housekeeping, since SegWit outputs are cheaper to spend than legacy ones. The transaction cannot tell the two apart, and we do not claim either.
One detail limits the quantum reading. These coins were never in the most exposed group. Pay-to-public-key (P2PK) outputs, used mainly in 2009 and 2010, show the public key from the moment they are created. None of the seven inputs was of that type, and the new bc1q addresses again hide their keys behind a hash until they are spent.
2013-era coins also moved in January and May, with no reason given
Saturday’s move is not the first of its kind in 2026. On 19 January, The Block reported that 909.38 BTC, worth about $84.6 million, moved after more than 13 years; it quoted bitcoin at $92,531 later that evening. Those coins had been gathered between December 2012 and April 2013 and went to a single new address, a move flagged by Arkham.
On 10 May, The Block reported a 500 BTC transfer, about $40.6 million with bitcoin near $81,721, from an address dormant since 27 November 2013. The coins went to a new bc1q address, according to Arkham data.
Both episodes followed the same path as this one: legacy coins from bitcoin’s early years moved to a modern address, with no statement from the owner. The Block gave no reason in either case, while noting in May that such transfers often come before a sale.
Bitcoin.com, citing btcparser.com, counts 774.17 BTC of long-dormant coins moved so far in October, against 5,401.84 BTC in September and 6,427.59 BTC in August. The outlet reads the slower October as a sign that long-term holders are waiting for a price direction. That is its interpretation, and the tallies are btcparser’s own; we did not check them.
What to watch on the 15 new addresses
The new outputs are public and can be followed on any block explorer from the transaction ID, d46a16cc…a61dc. A first spend into an address known to belong to an exchange would turn this into a sale story. Spends between the new addresses, or into a multisig script, would point to a custody reshuffle instead.
Until one of them moves, the record shows a holder of 471.10 BTC who signed a single, unhurried transaction on Saturday evening and has kept the coins in place since. btcparser.com’s count for the whole of October will show whether more coins from the same era follow.