Grin & Mimblewimble: Voldemort's Invisibility Cloak and the Privacy Frontier

1. A Spell Thrown Into the Research Channel

Grin's history dates the drop to August 1, 2016: a user named majorplayer logged into the #bitcoin-wizards IRC channel, posted a Tor-hosted text file, and disappeared. Andrew Poelstra's follow-up separately places an anonymous person using the name Tom Elvis Jedusor on a Bitcoin research IRC channel at about 04:30 UTC on August 2. The document itself was titled Mimblewimble, signed Tom Elvis Jedusor, and dated 19 July 2016. Its complaint was concrete: nearly 150 million Bitcoin transactions had to be replayed to confirm 4 million unspent outputs, while the transaction graph fed companies whose business was monitoring users.^[3][2][1]

hi, i have an idea for improving privacy in bitcoin. my friend who knows technology says this channel would have interest^[3]

— majorplayer

Grin's documentation notes that Tom Elvis Jedusor is the French name for Tom Riddle — Lord Voldemort — and Mimblewimble is Harry Potter's tongue-tying curse. The author explained the name in one line:^[3]

I call my creation Mimblewimble because it is used to prevent the blockchain from talking about all user's information^[1]

— Tom Elvis Jedusor

In October 2016, Andrew Poelstra published a follow-up paper, dated 6 October, that tightened the design. It estimated 15 gigabytes for Bitcoin-scale transaction history and claimed compact-chain data could fall below one megabyte, excluding the UTXO set, which it estimated above 100 gigabytes with rangeproofs. Within two weeks, the idea had a second pseudonym.^[2][3]

2. The Magic: No Amounts, No Addresses

The design starts by removing Bitcoin Script. Amounts become Pedersen commitments, C = r × G + v × H, which hide each value behind a secret blinding factor. Conserving value means inputs cover outputs plus the fee; it does not mean the full commitments must sum to zero. The original paper introduces a nonzero excess and a signature proving knowledge of its secret key. Grin’s technical guide includes signed kernel excesses and a kernel offset in the balance check, while rangeproofs rule out invalid negative amounts. Together these checks protect the supply without publishing payment amounts.^[1][13]

Grin replaces visible addresses with control of secret blinding factors. Cut-through then cancels a spent output against its matching input. It does not erase everything: unspent outputs and their rangeproofs, transaction kernels, and block headers remain necessary. The roughly 100-byte kernel cited by Grin is lasting overhead per transaction, not the size of all surviving chain data. Its documentation’s estimate of growth at roughly one quarter Bitcoin’s rate is an asymptotic project claim.^[4][13]

However, to achieve these goals, Mimblewimble cannot support a general-purpose scripting system such as that in Bitcoin.^[2]

— Andrew Poelstra

That was the price: Poelstra's paper warned that dropping script precludes zero-knowledge contingent payments, cross-chain atomic swaps, and micropayment channels in their Bitcoin form. Grin's developers later documented elliptic-curve routes back to multi-signatures, atomic swaps, timelocks, and Lightning-style channels.^[2][4]

3. Interactive Money and the Edges of the Cloak

Spending is a handshake. The original paper prescribes a sender–recipient "ritual" for jointly building each transaction, and Grin's FAQ confirms wallets must exchange data directly — though not simultaneously online, since the exchange can run over email or Signal.^[1][4]

The cloak has edges. Grin's documentation says a relay listener can identify which outputs a pending transaction spends before cut-through; it presents Dandelion as protection for originating IP addresses. Ivan Bogatyy reported tracing roughly 96 percent of Grin transactions using three sniffer nodes over several days in May 2019, reconstructing who paid whom while amounts remained protected. His write-up says Grin's developers knew such an attack was theoretically possible.^[4][11]

Beam offered a different boundary: contemporaneous reporting described an option for users to make transactions auditable. That is deliberate disclosure by the user, distinct from an observer reconstructing links during relay.^[7]

4. Beam First, Grin Twelve Days Later

Beam moved first: the company-built chain launched mainnet on January 3rd, 2019, with zero coins in circulation and — per Beam's tokenomics — no pre-mine or ICO. Twelve days later, on January 15th, Grin's genesis block was mined.^[6][3][7]

Beam's funding model was written into emission. Its official documentation scheduled 20 Treasury coins per block in year one and 10 through year five, intended to repay investors, incentivize the core team, and fund the Beam Foundation. It said the monthly distribution would be roughly 35 percent investors, 45 percent core team and advisors, and 20 percent Foundation, from a supply capped at 262,800,000 coins with Bitcoin-style halvings.^[6]

Grin ran the other way, as contemporaneous coverage framed it: "the philosophically pure cryptocurrency – community driven and funded, not premined and with no taxation on mining," set against Beam's company-controlled 20 percent. At the time of that January 7 report, Beam's hashrate had reached 1.68 MSol per second on Equihash. Both framings are journalism's, not adjudications.^[7]

Grin's manifesto was its emission: a new grin every second, forever — 60 grin per block on a 60-second block goal, with no maximum supply. Its stated rationale was a constant subsidy protecting mining over decades, with relative dilution below 10 percent after ten years. Those are project intentions, not audited outcomes. Months after launch, Ignotus Peverell — named for the wizard who received the Cloak of Invisibility — disappeared; contributors continued the work, some funded by donations.^[5][4][3]

5. Delistings, Extension Blocks, and a Cloak That Fit Other Shoulders

A September 23, 2019 report described planned exchange delistings: Upbit had scheduled six privacy coins for September 30, and OKEx Korea separately scheduled five for October 10, citing FATF guidance. Neither list included Grin or Beam. This report establishes the announcements and dates, not that either exchange had already completed them. Nor were these court judgments or a ban on the protocol.^[9]

Litecoin reached the 75 percent miner-signaling threshold on May 2, 2022, locking in MWEB before its activation later that May at block 2,257,920. Activated as a soft fork, the extension-block feature hides amounts for users who opt in; it does not provide complete privacy. Unlike the original interactive design, MWEB also supports noninteractive transactions. That June, Cointelegraph, citing 8BTC, reported that five major South Korean exchanges announced Litecoin delistings after flagging MWEB; its article directly documents only Upbit’s AML-law analysis.^[8][12]

In terms of fungibility and privacy, I believe MWEB gets you 90% there. For most people that's good enough. It's the difference between living in a glass house vs living in a house with windows. For people who need 100% privacy, they can live in a house with no windows.^[8]

— Charlie Lee

People carried the migration: MWEB's lead developer was David Burkett of Grin++, paid through community funding. Grin's repository still states the chain is live with mainnet, and Beam's documentation remains active. The core idea outlived its first two launches — an editorial interpretation, not a ledger fact.^[8][10][6]