The Bitcoin Policy Institute just published an open letter asking AI labs to give open-source bitcoin and digital asset defenders access to their strongest cybersecurity models.
The problem: "Open-source defenders today occupy the least privileged position in the AI security landscape."
Bitcoin secures over a trillion dollars in value, but the developers maintaining wallets, cryptographic libraries, and node software are getting blocked by guardrails when they try to use frontier AI for legitimate security work.
The letter warns that "nation-state adversaries, criminal syndicates, and sophisticated hackers will not wait for permission to deploy AI against critical open-source infrastructure."
Defenders are stuck using weaker open-weight alternatives while attackers face no such constraints.
BPI is asking for early access to the best cyber models, compute budgets for real security reviews, and secure environments for sensitive code analysis.
The coalition is massive: Coinbase, Block, Strategy, MARA, Blockstream, Chaincode, Brink, Ten31, Unchained, Trezor, Casa, Strike, River, OpenSats, and dozens more.
"Frontier AI could become one of the most powerful defensive technologies ever developed, but only if defenders get fair access."


Defenders Need the Frontier
An open letter to frontier AI labs: the defenders of open-source financial infrastructure need the frontier too, before attackers do.

