KYC Is Outdated Security Theater

Every company that collects your identity documents is making a bet. They're betting they won't get hacked, that their third-party vendors won't get hacked, and that none of their own staff will sell you out. They're betting that the data they collected "for your security" won't be used against you. History says they lose that bet.

The argument for KYC (Know Your Customer) is always the same: "We need your government ID, your selfie, your home address, and your phone number... for security." The logic runs backwards.

Collecting sensitive identity documents makes users targets. It builds honeypots that hackers, insiders and state actors go looking for. When those honeypots get breached, and they do, the people who handed over their data pay the price.

The Breach Hall of Fame

Every case below happened, and every one started with a company holding identity documents.

Discord (October 2025)

Discord collected government IDs for age verification. A third-party vendor got compromised. Result: 70,000 government ID photos exposed. The hackers claimed to have over 2.1 million passport and driver's license images and demanded ransom.

The Electronic Frontier Foundation put it bluntly at the start of the year: "Mandatory age verification tools are surveillance systems that threaten everyone’s rights to speech and privacy, and introduce more harm than they seek to combat."

Discord's defense was "it wasn't us, it was a third party," as if that matters to the people whose passports are now on hacker forums.

Ledger (2020): "When it got Physical"

Ledger, the cryptocurrency hardware wallet company, collected customer shipping addresses to fulfil orders. Reasonable enough, until they got breached.

1.1 million email addresses and 272,000 records with full names, phone numbers, and home addresses leaked, and the data was dumped publicly online. What followed:

The breach cost Ledger over $33 million in remediation. Nobody can put a number on what it cost users, and I'd bet some are still being targeted years later.

Coinbase (2024-2025)

In December 2024, hackers bribed a Coinbase employee in India to leak sensitive customer data. 70,000 customers were affected, at an estimated cost of up to $400 million.

The stolen data included government-issued IDs, home addresses and other KYC documents. A former customer service agent was arrested, but the data is already out there. Arresting one agent doesn't put it back.

Nobody needed a zero-day for this. Someone with access took a bribe, and every company holding KYC data has employees with access.

MobiKwik (2021): "The Largest KYC Leak in History"

A white hat hacker discovered what they called "probably the largest KYC data leak in history": 8.2 terabytes of data from the Indian mobile payment app.

The haul included:

  • ID scans, passports, Aadhar cards, PAN cards
  • Selfies used for identity verification
  • Emails, phone numbers, addresses
  • 99 million records with GPS locations and apps installed
  • 3 million merchant KYC documents

All of it was for sale on hacking forums for 1.5 BTC.

Hetzner (2017, 2018, and counting)

The hosting company, which requires extensive KYC, was breached twice in a single year. Customer data exposed included:

  • Names, email addresses, phone numbers
  • Physical addresses
  • Identity numbers
  • Bank account numbers
  • VAT numbers

Their breach history goes back to at least 2013, and they still require all that data from new customers.

The List Goes On

More KYC Disasters:

• Transak (2024): 92,554 users' government IDs and selfies stolen
  via phishing attack on ONE employee
  Source

• Indian Financial Institutions (2025): 500GB of KYC data exposed
  through a misconfigured storage bucket
  Source

• UK KYC Documents (2025): 1GB of selfies, IDs, passports,
  driver's licenses dumped on dark web forums
  Source

• Signzy (2024): Major KYC provider breached, customer data
  from banks appeared on dark web
  Source

• NCX Exchange (2025): 2+ million records leaked including
  KYC documents and internal keys
  Source

• Binance (2019): 10,000+ KYC photos allegedly stolen,
  hacker demanded 300 BTC ransom
  Source

• MOVEit Hack (2023): 2.85 million KYC records stolen
  from 60 banks in a single attack
  Source

Look at that list long enough and the question stops being whether your KYC data will leak and becomes when.

Destroying the "Security" Argument

The standard defense of KYC goes something like this: "We collect this data to protect you from fraud and ensure platform security."

Compare what's claimed with what actually happens:

Claimed Purpose          | Actual Result
-------------------------|------------------------------------------
"Prevent fraud"          | Creates fraud vectors via data breaches
"Account security"       | Enables account takeover via leaked data
"Verify identity"        | Identities get stolen and misused
"Regulatory compliance"  | User data becomes regulatory liability
"Protect users"          | Users get extorted, phished, robbed

The Pattern:
1. Company collects KYC data "for security"
2. Company stores KYC data (or gives it to third party)
3. Breach occurs (hack, insider threat, misconfiguration)
4. Users' "secure" data is now in criminal hands
5. Users face identity theft, phishing, physical threats
6. Company issues apology, offers credit monitoring
7. Repeat

In 2024, finance overtook healthcare as the most hacked industry, accounting for about 152 million data breach records. More than 60% of compliance officers now report data breaches as a top stressor.

On top of that, KYC doesn't stop sophisticated criminals. 404 Media reported that half of identity checks can now be bypassed with AI-generated fake IDs.

KYC catches honest people. Criminals use synthetic identities, stolen documents and AI-generated faces, so the system punishes legitimate users and barely slows down the bad actors.

Destroying the "Law Enforcement" Argument

The second argument is always: "If we didn't collect KYC, what would we do when law enforcement asks for customer information?"

I brought this up in a previous post and I'll bring it up again, because the Mullvad story answers it.

In April 2023, Swedish police raided Mullvad VPN's offices with a search warrant. They wanted user data, customer information, connection logs, anything they could get.

They left empty-handed.

Mullvad cooperated and destroyed nothing. There was simply no data to give. A Mullvad account is a randomly generated number with no email, no name and no records behind it, so there was nothing to hand over. You can't leak what you never collected.

Companies like Hetzner and Discord don't have to demand personal information from their customers. If abuse and bots are a problem on their platform, the fixes are simple:

  • Abuse on Hetzner servers or IP ranges: process the abuse reports and, if needed, close the customer's account.
  • Bots on Discord: encourage users to report spam and pay humans to review those reports. And don't come complaining when your automated moderation AI starts banning people over false reports. I did say pay humans to review them, but that's a story for another day.

Law enforcement has plenty of tools. It doesn't need every cloud provider to become an identity honeypot. The companies that require extensive KYC mostly aren't doing it because law enforcement demands it. My guess is they can't be arsed to update their platforms, even after it's been shown time and again that KYC is a huge liability and nearly useless.

The Honeypot Problem

KYC proponents never deal with this:

Every database of identity documents is a target. The more valuable the data, the bigger the target. Government IDs, passports, and biometric selfies are the most valuable data that exists.

When you require KYC, you're painting a bullseye on your infrastructure. Hackers, insiders, and state actors all want what you're holding.

The threat isn't only external. The Coinbase breach happened because someone got bribed: the data existed, so it could be sold. The 2025 Verizon Data Breach Investigations Report found that 30% of breaches now involve third-party vendors.

You can have the best security team in the world, encrypt everything and audit constantly, and you still can't eliminate:

  • Insider threats (employees can be bribed, blackmailed, or go rogue)
  • Third-party vendor risks (you're only as secure as your weakest vendor)
  • Zero-day exploits (new vulnerabilities are discovered constantly)
  • Social engineering (people can be talked into almost anything)
  • State-level actors (some adversaries have unlimited resources)

The only way to be sure you never leak identity documents is to not collect them in the first place.

The Real Consequences

This is what happens to people when their KYC data leaks:

From the Ledger breach:

"We know you have Bitcoin. We know where you live.
Pay us $800 or we're coming to your house."

(Actual extortion message sent to Ledger customers)

From leaked KYC data in general:

• Identity theft (opening fraudulent accounts in your name)
• SIM swapping attacks (taking over your phone number)
• Targeted phishing (attackers know exactly what services you use)
• Physical robbery (they know you have crypto AND where you live)
• Synthetic identity fraud (your documents used to create fake identities)
• Bypassing KYC on other platforms (using YOUR stolen docs)
• Long-term surveillance (your data never expires)

"Wrench attacks", physical robberies targeting cryptocurrency holders, are on the rise in 2025. Criminals use leaked KYC data to pick targets and find their homes. And you can't change your face or your fingerprints the way you change a password.

When a password leaks, you change it. When your government ID leaks, you stay exposed for life.

What Real Security Looks Like

Privacy-respecting services have already shown the alternative works:

The No-KYC Architecture:

Authentication:
- Random credential (account number, API key, etc.)
- No email (can't be phished, can't be leaked)
- No phone (can't be SIM swapped)
- No identity documents (can't be stolen)

What gets stored:
- Credential hash
- Account balance/status
- Active services

What doesn't get stored:
- Real name
- Email address
- Phone number
- Physical address
- Government ID
- Selfies
- IP logs
- Usage analytics

Result when breached:
- Random hashed strings leak
- No identity connection
- Nothing to extort with
- No physical threats possible

Mullvad has been operating this way for years. When they got raided, what protected users was the design of the system itself, with no heroics required from their security team.

The Trade-Off They Won't Tell You About

Every service that requires KYC is making a trade-off on your behalf:

"We're trading your long-term security for our short-term convenience."

Email has become a pseudo-standard for registering on online platforms, and even after thousands of data breaches most platforms haven't learned their lesson. It's easier to demand a phone number than to implement proper 2FA, and easier to collect government IDs than to accept the legal complexity of not having them. What's easier for the company is often worse for the user.

Credential-based systems have trade-offs too. Lose your credential and you lose your account, with no recovery process and no "verify your identity" workflow. That's the cost of actual privacy.

Now consider the alternative. You give up your government ID, and when it leaks you deal with identity theft for the rest of your life, get extortion emails, and wonder if someone's going to show up at your door.

Stop Handing Over Your Passport

KYC is outdated security theater that makes users less safe.
Every database of identity documents will eventually breach.
The only winning move is not to collect the data.

Companies that want your government ID, your selfie and your home address "for security" create the very vulnerabilities they claim to prevent. They build honeypots and call it compliance.

If a service asks for your passport and has no legal reason to, take your business somewhere that doesn't. Otherwise you're just waiting for your passport to show up on a hacker forum.

Further reading

Deploy
this

Compare
providers

Best Anonymous Dedicated Servers (2026): The Short List That Does Not Want Your Passport Dedicated servers are where privacy hosting gets thin: most of the market wants an identity before it hands… Best VPS for Tor Relays and Bridges (2026): Who Allows What, According to Their Own Policies and the Tor Project Ten hosts for running a Tor relay, bridge or exit, ranked by their written policies and the Tor Project… Best Offshore VPS Providers (2026): Jurisdictions, DMCA Policies and Signup Rules Twelve offshore VPS providers ranked by what "offshore" buys you: the legal jurisdiction your data sits in… Best Crypto VPS Hosting (2026): Who Takes Bitcoin, Who Processes It Themselves, and Who Still Wants Your Passport Fifteen VPS providers that accept cryptocurrency, ranked by what the crypto actually buys you: whether… Best VPS Hosting You Can Pay With Monero (2026) Ten VPS providers that take Monero, ranked by how they handle it: own node or third-party gateway, what signup… Best Confidential VPS Providers (2026): Encrypted Memory, Attestation, and Who Still Needs Your Name Seven ways to rent a virtual machine whose memory the operator cannot read, from a no-KYC host with… Best No-KYC VPS Providers (2026): Eleven Hosts That Skip the Identity Check Eleven VPS providers that will rent you a server without a passport, a selfie or a phone number, ranked by… Best VPS for Trading: What Actually Matters Beyond Price Most "best VPS for trading" lists rank by price and unverified latency numbers. What matters: latency measured… Best Anonymous VPS Hosting Providers Six anonymous VPS providers compared and re-verified in September 2026: signup requirements, logging, Monero…

More
articles

Support

Questions before you buy? Ask here. You do not need an account: your first message creates one, and you get its credential to come back for our reply.