Before Bitcoin: The Failed Attempts at Digital Cash

Bc6m...o9Qb

8 Sept 2026

Bitcoin didn't appear out of nowhere in 2008. For nearly two decades before Satoshi Nakamoto's whitepaper, a small, overlapping group of cryptographers and cypherpunks kept trying to build digital cash, and kept failing, for a handful of very different reasons. Each failure quietly solved one piece of the puzzle while missing another. Here are five of them, and what each one got right before it went wrong.

The currency that turned down Bill Gates (DigiCash)

In 1989, cryptographer David Chaum founded DigiCash in Amsterdam, built on a genuinely groundbreaking 1982 paper of his called "Blind Signatures for Untraceable Payments." The idea was real digital cash, anonymous, cryptographically secured electronic coins that banks would issue and websites could accept, years before most people had ever used the internet to buy anything at all. DigiCash signed real partnerships with major banks, including Deutsche Bank and the Mark Twain Bank in the US, and its ecash product launched in 1995 to genuine industry buzz.
Then came the deal that, in hindsight, might have changed everything: Microsoft reportedly wanted to bundle ecash directly into every copy of Windows 95. Chaum held out for more money than Microsoft was willing to pay, and Bill Gates walked away. Without that kind of massive built-in distribution, DigiCash was stuck in a chicken-and-egg trap, not enough merchants accepted it to attract users, and not enough users had it to attract merchants. The company filed for bankruptcy in 1998.
Why it failed: the cryptography worked perfectly. What DigiCash never solved was distribution and adoption, and unlike a purely peer-to-peer system, it still depended on banks choosing to issue it and merchants choosing to accept it, which meant its fate rested on a handful of corporate decisions rather than the technology itself.

Digital gold that attracted criminals faster than investors (e-gold)

Launched in 1996 by oncologist-turned-entrepreneur Douglas Jackson, e-gold let users hold and transfer value backed by real, physically-stored gold reserves, no bank account required. It grew enormously: by the mid-2000s it had over 3 million accounts across more than 165 countries, settling more than $2 billion a year in transactions.
That same borderless, low-verification design that fueled its growth also made it the payment method of choice for a very different customer base. In 2007, the US Department of Justice indicted e-gold and its founders, alleging the platform had become a preferred tool for money launderers, identity thieves, and most damningly, child pornography distributors. Jackson and his co-founders pleaded guilty in 2008 to money laundering and operating an unlicensed money transmitting business, and the service was effectively shut down.
Why it failed: e-gold solved the technical problem of digital, gold-backed value transfer just fine. What it never solved was the reality that a payment system open to literally anyone, anywhere, with minimal oversight, becomes irresistible to criminals long before it becomes trusted by regulators, and there was still a company, and a person, that law enforcement could actually indict.

A spam filter that accidentally invented Bitcoin's engine (Hashcash)

Not every entry on this list was even trying to build money. In 1997, cryptographer Adam Back proposed Hashcash to solve a completely different problem: email spam. His idea was to force an email sender's computer to do a small amount of costly computational work (e.g. solving a puzzle that took real, deliberate effort) before a message could be sent. Sending one email would cost almost nothing in computing time. Sending ten million spam emails would suddenly cost a meaningful amount of real computing power.
Hashcash never caught on widely as a spam filter. But the underlying trick (make cheating computationally expensive, while making verification instant) turned out to be exactly the missing ingredient for digital money. Satoshi Nakamoto cited Hashcash directly in the Bitcoin whitepaper, and Bitcoin's entire mining process is, structurally, Hashcash's puzzle repurposed to secure a ledger instead of a mailbox.
Why it matters: Hashcash didn't fail at solving spam so much as succeed at solving something nobody had asked it to solve yet. It's the clearest example on this list of a good idea simply waiting for the right problem to attach itself to.

The proposal that was almost Bitcoin, a decade early (b-money)

In 1998, on the Cypherpunks mailing list, a computer scientist named Wei Dai published a short proposal called b-money, a scheme for anonymous, distributed electronic cash, secured by solving computational puzzles and maintained through a shared, replicated ledger. Read today, it's eerie how much of Bitcoin's actual design is already sitting right there: money created through provable computational effort, and a public record everyone maintains together instead of one central party.
Dai never built it. The proposal was detailed on concept but genuinely light on the hard engineering question that had sunk every attempt before it, exactly how do you get a decentralized network of strangers to agree on one canonical order of transactions, without anyone cheating? Ten years later, Satoshi Nakamoto explicitly credited b-money in the Bitcoin whitepaper's opening lines, and by Dai's own account, Satoshi hadn't even read his proposal before independently arriving at similar ideas, b-money was cited afterward, once the parallels became obvious.
Why it failed to launch: b-money nailed the vision almost perfectly. What it lacked was the specific mechanism, the actual consensus algorithm, that could make a real network of untrusted strangers agree on the truth in practice, not just on paper.

The blueprint that was one piece short of Bitcoin (Bit Gold)

Also in 1998, cryptographer and legal scholar Nick Szabo began circulating an idea he called Bit Gold among the same small circle of cypherpunks, later publishing a full description in 2005. Bit Gold combined proof-of-work puzzles (echoing Hashcash) with a public registry that recorded which cryptographic public key owned which solved puzzle, a chain of timestamped, linked proofs strikingly similar to what would become Bitcoin's blockchain.
Szabo never built a working implementation. The design's Achilles' heel was the same one that had doomed every attempt before it: the registry needed a set of trusted "club member" servers to agree on who owned what, and that requirement for trusted coordination reopened exactly the door decentralized cash was supposed to close, the double-spending problem was never fully solved. One week after Szabo's fuller 2005 writeup gained attention, the timing overlapped with early cypherpunk discussions that would soon lead to Bitcoin, though Bit Gold, unlike b-money and Hashcash, was conspicuously never cited by name in Satoshi's original whitepaper, a small mystery that still fuels speculation about who Satoshi actually was.
Why it fell short: Bit Gold got remarkably close to Bitcoin's actual architecture (proof-of-work, public keys, a linked chain of records), but still needed a trusted group of servers to keep everyone honest, which meant it never actually removed the middleman it was designed to eliminate.

What Bitcoin actually did differently

Look at the five failures side by side, and Bitcoin's real innovation comes into focus, and it's not any single piece of technology, since every piece already existed. DigiCash proved digital cash could be cryptographically real. e-gold proved people would use it at scale, and proved exactly why unrestrained scale invites regulators. Hashcash proved you could make cheating computationally expensive. b-money sketched almost the entire vision. Bit Gold got closest of all, missing only one piece: a way to reach agreement without trusting anyone.
Bitcoin's actual contribution was gluing these pieces together with one final idea none of them had: a way for a fully decentralized network of total strangers to agree on a single, tamper-proof order of transactions, with no club of trusted servers anywhere in the design. Every failure on this list was one honest, unsolved problem away from being Bitcoin. It just took until 2008 for someone to solve the last one.