Binance founder Changpeng Zhao, widely known as CZ, has warned that the crypto community is focused on the wrong threat when it worries about artificial intelligence breaking Bitcoin’s encryption. His message: the bigger risk to crypto holders is already here, and it has nothing to do with cracking cryptographic code.
CZ argued that AI does not need to break Bitcoin’s underlying cryptography to cause serious harm to crypto users. The concern about AI eventually defeating the math that secures Bitcoin misses the point of how most real attacks actually happen. For related coverage, see Grayscale Lets Eligible Investors Swap Crypto for ETF Shares.
Bitcoin’s cryptographic security, the system of complex math problems that protects wallets and transactions, is extraordinarily difficult to crack with today’s computers or near-future AI systems. Researchers widely consider it safe for the foreseeable future. But cryptography being unbroken does not mean crypto holders are safe.
The Real Risk: Human Behavior, Not Math
The bigger threat CZ pointed to is the human layer of crypto security, not the technical one. Most successful crypto thefts do not involve breaking encryption. They involve tricking people into giving up access themselves.
AI makes social engineering attacks, where criminals impersonate trusted contacts, create fake customer support agents, or generate convincing phishing messages, far more scalable and believable. A scammer using AI can now produce personalized, fluent, convincing fraud attempts at a volume and quality that was previously impossible.
This concern is not theoretical. South Korean regulators reported that breaches exposed over 100,000 data items from crypto firms, the kind of personal information that feeds targeted phishing and impersonation attacks. When attackers know who you are, what exchange you use, and what your email address is, AI-powered fraud becomes dramatically more effective.
The lesson is that the weakest point in most crypto security setups is not an algorithm. It is the person holding the keys.
Why This Matters More Than Quantum or AI Hacking Headlines
Headlines about AI or quantum computers eventually cracking Bitcoin tend to generate fear and clicks. But security researchers consistently point out that such an attack would require computing resources so far beyond what exists today that it remains a distant and manageable theoretical concern.
Meanwhile, AI-assisted phishing, deepfake impersonation of project founders or exchange staff, and automated scam operations are growing threats right now. If an attacker can convince a user to approve a transaction, share a seed phrase (the master password to a crypto wallet), or connect their wallet to a fraudulent site, no amount of cryptographic strength protects that user.
Exchanges have responded to this reality by investing in identity verification and anti-phishing systems. Kraken’s unified wallet approach, for example, consolidates holdings in ways that affect how users interact with security controls, making the design of user-facing security more consequential, not less.
What Crypto Holders Should Actually Watch For
CZ’s warning points toward a practical checklist for anyone holding crypto, whether on a major exchange or in a self-custody wallet.
Seed phrases should never be entered online, shared with anyone, or stored in photos or cloud documents. No legitimate exchange, project, or support team will ever ask for them. AI-generated messages can now closely mimic the language and branding of real platforms, so users should verify any urgent request through official channels before acting.
Hardware wallets, physical devices that store private keys offline, protect against most software-based attacks. Two-factor authentication using an app rather than SMS provides a meaningful layer of protection against SIM-swapping, a common attack where criminals redirect a phone number to steal verification codes.
Regulatory conversations are also shifting toward this threat model. Ripple’s push for changes to SEC crypto rules and Hyperliquid’s policy recommendations to the SEC both touch on how the regulatory environment shapes platform security standards, including how exchanges handle user data and authentication.
CZ’s point is straightforward: worrying about AI eventually breaking Bitcoin’s math while ignoring AI-powered fraud happening today is like worrying about a bank vault being cracked by a future supercomputer while ignoring the person walking out the front door with your keys. The biggest crypto security risks are already here. Protecting against them requires basic habits, not waiting on cryptography researchers.
Additional source references: source document 1, source document 2.
Disclaimer: This article is for informational purposes only and does not constitute financial or investment advice. Cryptocurrency and digital asset markets carry significant risk. Always do your own research before making decisions.
