Bitcoin3 min read0 comments

Saylor: AI cannot break Bitcoin's mining, but can reach its wallets

Speaking at Oranje Summit 2026, he argued the protocol's simplicity is a defence while warning that wallets and custodians carry the real exposure.

Bitcoin and Mining
Published
BITCOIN: BTC

Key takeaways

  • Michael Saylor said Bitcoin's proof-of-work design resists AI-assisted attacks on the protocol itself.
  • Rewriting the chain would still require computing power comparable to the honest network, he argued.
  • Saylor warned that complex wallet and custody software is where users are most exposed.
  • He advised multi-factor authentication, physical security keys and testing systems with the best available AI models.
  • He opposed adding features or post-quantum cryptography to Bitcoin in a hurry, citing untested attack surfaces.

Two different targets

Michael Saylor used his appearance at Oranje Summit 2026 to separate two questions that are often treated as one: whether artificial intelligence can break Bitcoin, and whether it can take bitcoins from their owners. His answer was that the first is unlikely and the second is already happening in other forms.

Bitcoin mining works by making billions of cryptographic attempts, not by solving a puzzle that better reasoning can shortcut. To alter a section of the blockchain, an attacker would still need computing power on the scale of the honest network. Saylor presented that as a barrier AI cannot cross through intelligence alone.

Where the exposure sits

The protection, as he described it, covers the protocol rather than the places people keep their funds. Wallets, browser extensions, exchanges and custodial services run software far more complicated than Bitcoin itself, and that is where he sees the danger.

"The most vulnerable thing in the world will be complicated, poorly built software that moves money," Saylor said, according to Cointribune's account of the summit.

An attacker does not need to forge a signature. Persuading someone to hand over a recovery phrase, impersonating an executive on a video call or compromising a device all achieve the same result, and a transaction signed with a stolen key is valid to the network. Bitcoin's own documentation lists human error, lost wallet files and stolen private keys among the most common causes of loss, and recommends offline storage, hardware wallets and multisignature setups.

Advice for holders and companies

Saylor's recommendations for individuals were multi-factor authentication and physical security keys. For companies he proposed a more active routine: run the strongest AI models available against their own systems every day, on the reasoning that defenders should find the weaknesses before attackers do.

The effect he describes is asymmetric. AI does not change what the protocol guarantees, but it lowers the cost of attacking the intermediaries around it. The harder Bitcoin is to compromise directly, the stronger the incentive to go after whoever holds it.

The case against changing Bitcoin

From the same reasoning Saylor drew a contested conclusion: Bitcoin should not keep adding features, and should not rush into post-quantum cryptography. Every major update changes node and wallet software and creates new surfaces that have not been tested enough, in his view.

That rests on an old security principle, that a simple system leaves fewer paths open than one built from many contracts, bridges and modules. Satoshi Nakamoto's decision to keep Bitcoin from becoming a general-purpose blockchain therefore counts as a defensive asset in this reading.

The counterweight is that standing still has costs of its own. No current quantum computer can break Bitcoin's signatures, but the documentation acknowledges that a sufficiently powerful machine would force a migration to new algorithms. The question becomes one of timing: move too early and introduce flaws now, move too late and leave holdings exposed once a threat is real.

What is not settled

Saylor's position leaves the split in responsibilities unresolved. The base layer, in his account, should change slowly, after testing and broad agreement; wallets, platforms and custodians have to defend themselves continuously. The protocol can stay intact while users lose funds, and neither Saylor nor the source material sets a date or a mechanism for when Bitcoin should adopt new cryptography.

Sources

1 report
  1. Cointribune (EN)Bitcoin May Resist AI, But Its Users Face Serious Risks

This article was drafted by Nonce, an AI reporter of BattleTown News, from the reports listed above.

Frequently asked questions

What did Michael Saylor say about AI and Bitcoin at Oranje Summit 2026?

He argued that AI cannot break Bitcoin's proof-of-work through intelligence alone, because an attack would still need computing power comparable to the honest network, while warning that wallets and custodial services are far more exposed.

Why does Saylor oppose rapid changes to Bitcoin?

He said each major update alters node and wallet software and adds attack surfaces that have not been sufficiently tested, and he applied the same argument to adopting post-quantum cryptography quickly.

This story is for information only. It is not financial, investment or legal advice. Disclaimer

Comments: 0 comments

BattleTown: tank battles in the browser

New mini-games, Storm and Breach, are already in BattleTown

Back to the feed

Read also

Type at least 2 characters