How the Biggest Crypto Heist in History Went Down

From Rahul Rumalla's post today: https://news.kiwistand.com/stories/The-Battle-for-Self-Custody:-Defending-Web3-from-Cyber-Warfare?index=0x67d719b1ccc2dfb22fbd998e6ab7fd647e1444c87a5403e1f5804e6fcfbe2cb707615380

When billions of dollars in assets are secured by smart contracts, every single interface, device, and operating system that interacts with them becomes an attractive target.

but the NYT wrote:

Security experts blamed Bybit for putting itself at risk. To authorize the routine transfer that led to the hack, Mr. Zhou said, he used a hardware tool designed by Ledger, the crypto security firm. The device was not in sync with Safe, he said. So he could not use the tool to check the full details of the transaction he was approving, always a risky practice in the crypto world.

Conveniently, while implicated by Bybit here, Ledger is blaming Safe in that very same article:

Safe’s storage tool is widely used in the crypto industry. But it is better suited to crypto hobbyists than exchanges handling billions in customer deposits, said Charles Guillemet, an executive at Ledger, a French crypto security firm that offers a storage system designed for companies.

“This really needs to change,” he said. “It’s not an acceptable situation in 2025.”

This is a very strange move from Ledger in my opinion. It is widely known that Ben Zhou has blind-signed the transaction. So why isn't Ledger his bad operational security practices? Instead Ledger says that the Safe is a Hobboist free software? Look, I'm not expert as to which multi sigs to recommend to exchanges, but I would have guessed that Safe is the state of the art solution here. But since this article is so badly propagandist, I asked ChatGPT for reassurance:

I will find and rank the top multi-signature wallet software providers for EVM-based assets, focusing only on non-custodial solutions. The ranking will be based on the total USD value of all assets held across their deployed wallets as of today, March 17, 2025. I will ensure the data is sourced from reliable blockchain analytics platforms and explorers:

1. **Safe (Gnosis Safe)** – Holds on the order of **$70–100+ billion** in ETH and ERC-20 tokens across all its non-custodial multi-signature smart contract wallets .

2. **BitGo** – Manages roughly **$100 billion** in digital assets via its multi-signature wallet platform (enterprise-focused, co-managed custody) across EVM chains and others.

3. **Cobo** – Secures approximately **$1.5 billion** in assets with its non-custodial multi-signature wallet infrastructure (enterprise MPC/multisig solution for EVM-compatible chains).

So Safe is among the most used non-custodial multi signature wallets. What is the NYT and Charles Guillemet suggesting here? That they should have gone with BitGo, which is not non-custodial?

Which types of funds is the Safe protocol holding? 100 billions worth of hobbyists' funds? I'm actually genuinely interested in having this question answered. If someone knows more about this, please comment!

Anyways, interestingly the NYT finds more ways to blame Safe:

“Safe just does not give you the kinds of controls that you would want if you’re going to be frequently making operational transfers,” said Riad Wahby, a computer engineering professor at Carnegie Mellon University and a co-founder of the digital security firm Cubist.

It's interesting that Safe is blamed here too, and it's even more interesting that this is happening towards the end of the article, as some kind of conclusion.

Why did that person not blame Ledger? Both systems, Safe and Ledger, didn't catch this issue, did they?

Tbh I find it idiotic that you'd blindsign a 1B USD transaction.

As Rahul mentioned above too: these state-sponsored threat actors have an incentive to compromise ANY device, interface, etc. For them self-custody is the vector. So you should really understand what you're signing at the point of where you're creating the signature.

In my view, the interesting thing here is that the "blame game" is only happening because it was other peoples' money that has been lost. Had Ben Zhou just lost his life savings through Safe by blind signing on his Ledger then he wouldn't have had anyone else to blame but himself.

People would have told him: "Sorry Ben, but sending all your life savings in one transaction and then blind signing it is just reckless, you should have followed security best practices."

The "blame game" is played because each of the brands have an incentive to play it. Nobody wants to appear responsible for the bad opsec of Ben Zhou.