[m00dy]

I can tell one of the ways that is missing in the article.

Let's say, you have a lot of Bitcoins and your buddy is a bitcoin miner. You craft your transaction such a way that you put all your coins as transaction fee. You send your transaction only to your buddy. Your buddy picks it up and solves the puzzle afterwards. Fees will be converted to brand new coins.

[flotzam]

Brand new coins, but wouldn't they still look suspicious if the block's coinbase is outlandishly large?

[Taek]

Also you'd be able to see which transaction paid the large fee and what the history is.

Might go unnoticed if the original coins weren't suspicious, but if an investigator is already looking at the original transaction because they suspect it was involved in crime, this type of jump is unlikely to throw them off the trail.

[aqme28]

That used to be common before wallets got better. People would accidentally set the fee way too high every once in a while.

[vmception]

It will literally be international news (in crypto circles) when a transaction like that gets mined.

People frantically contact all the mining pools to see who mined it and if they will return the funds to the sending address.

This has happened many times and they usually do return it, because people have nearly universal consensus that it was a mistake.

Kind of not a great way because it is too conspicuous.

There was one time this happened that was interesting and intended to be conspicuous:

Some hackers got access to an exchange, but the exchange had some pretty good security and would not let them withdraw large amounts, but the hackers could set the transaction fee. So they started burning all the exchange's money by distributing them to miners with this high transaction fee, to let the exchange know they were serious and needed their demands met.

Could they have coordinated with a miner and nobody would be the wiser? Sure.

[latchkey]

This is effectively MEV on ETH. Happens all the time.

[mgerullis]

Is it possible to send out transactions without sending them to the mempool?

[alcio]

yes, you can email it to your miner buddy.

it's possible to spot such transactions if they violate transaction forwarding rules (aka standardness rules) but not consensus rules. for example, a transaction greater than 100kB is not standard but still valid.

[b10c]

I've build a tool to detect differences between _my local_ mempool and what miners include in their block (there will always be slight differences). This is primarily intended to detect censorship, but can also detect transactions that never entered _my_ mempool.

See https://miningpool.observer

[JefffJohn555]

Brilliant! I remember thinking about this problem a few years back: what stops miners of a blockchain just ignoring transactions/anything from certain entities. So it's good that there exists a way to track such behaviour, if it is occurring.

[martinko]

Economics. If miners consistently ignore certain transactions, there is space for new miners to enter the market and earn super-normal profit. After the next difficulty adjustment cycle, the old, censoring miners may be priced out.

[neb_b]

This is really neat. Have you thought about trying to bundle it as an app that can be downloaded on umbrel?

[b10c]

Yes! Will be on umbrel eventually. Currently needs a Bitcoin Core build from the master branch. The features use should be in the upcoming release. Then Umbrel!

[neb_b]

Looking forward to it!

[mgerullis]

Yeah but wouldn’t it require some kind of fork of the BTC software (not chain) to actually include this transaction without sending it to everyone else?

[walls]

The transaction is valid to all miners, it just isn't shared with the network until your buddy finds a block that includes it.

[Taek]

Law enforcement watches the mempool, they can spot such transactions anyway because they know it didn't ever show up in their mempool logs.

[mgerullis]

Ok but is it really illegal? I doubt there’s a law on who can send who private BTC transactions? Or is it actually possible to mark this as laundering?

[Taek]

There's nothing illegal here. The goal of privacy is to avoid giving anyone a reason to suspect something. In the case of money laundering specifically, you want to turn money that looks extremely suspicious into money that doesn't look suspicious at all.

Little tiny 'gotchas' aren't going to save you from a determined investigator. They are trying to follow a trail of evidence so they can produce more evidence. What you need is a clean break, so that the investigator hits a dead end and has no productive leads they can follow.

[aqme28]

There will always be discrepancies due to e.g. latency of the network.

Just pretend you sent that transaction a millisecond before it was mined.

[vbezhenar]

What's the point? It's not hidden after block is mined. You would hide your trace if blockchain analysis tools did not account for that use-case, but that's probably a known method.

[exo762]

You can still trace the money by looking at the block explorer. All you did is thrown them in coinbase address. This is no different from coinjoin.

[koheripbal]

Your "buddy" would have to be a large mining pool for this to work, and even then it would be noticeable on the blockchain that this happened.

Worst idea to launder ever.

[crispyporkbites]

I don’t think that works at all? Your buddy would have to mine that specific block to get the reward, which is based entirely on chance. You’d also have to spend as much on the POW electricity and hardware to have a chance of getting reward anyway.

[Aissen]

If you know you mine blocks regularly (e.g once per day or week), it's entirely doable. It's the miners who chose what transactions go into a block. In this scheme, the transaction is "secret", only one miner received it.

[goldenkey]

The transaction isn't secret after it gets mined in the block and added to everyone's main chain. The only use of this would be to conceal one's IP for the transaction broadcast but it's really not necessary anyway because the bitcoin mesh network does a repeating broadcast from all peers to all peers, of all transactions in the mempool. Just use a secure connection and only broadcast to a single peer that you can verify to be a standard "dumb client." Odds are they won't have any logging of IPs and will rebroadcast the transaction, which will propagate it to the entire network.

[dopidopHN]

Yeah unless your buddy is a mining pool, I don’t follow how that would work as well.

[Aissen]

Yup, for Bitcoin it would mean having the mining power of a big pool, indeed.

[Retric]

No, just a secret block that isn’t broadcast to everyone. 1 block a week is plenty to pull this off and that’s just 1/1,000th of the worlds mining power. Unless you get really unlucky and the block fails to enter the block chain letting someone else gets credit for the transaction.

[Aissen]

I'm curious what it's like, so here is a small calculation. Some metrics:

- 144 blocks per day are mined on average

- the current network hashrate is 145M TH/s

- a 100 TH/s rig is about $10k.

The investment to be able to have full control of mining one block on average, without electricity, internet and storage :

- per week: you'd need 143k TH/s (145M / (144 * 7)), so about $14M of investment in just the mining rig (provided you can buy it all)

- per month: you'd need 33k TH/s, so $3M of mining rig investment.

- per year: you'd need 2.7k TH/s, so about $270k in mining rig equipment.

Of course, there are a lot of variables here (e.g hashrate is highly variable), but this gives a general idea.

All this for a "washing" method that heavily implicates the miner: the address of the new coins is still known, it's not really "clean", just an unusual transaction.

[latchkey]

Now run the same numbers for ETH.

[Taek]

If you're really patient (and many big time criminals are), you only need to mine once every few months, maybe once a year or less. That's still hundreds of thousands of dollars of mining equipment, but well in reach for many.