According to CoinMeta, over thirty Bitcoin and crypto companies have requested access to the largest AI laboratory to allow open-source security researchers early access to their most powerful models. They believe that the tools used by those responsible for protecting infrastructure worth trillions of dollars are far inferior to those used by attackers. The letter was organized by the Bitcoin Policy Institute, and signatories include Coinbase, Block, BitGo, as well as non-profit development funds such as Brink, Chaincode, and Btrust. The letter points out that Bitcoin core developers are unable to access the programs run by trusted security partners in the laboratory, and when they turn to publicly available models, security filters hinder their ability to discover vulnerabilities, forcing them to use less powerful open-weight models. The signatories request five things: early access to the most powerful network security models, sufficient computing resources, a secure environment for checking private code, eligibility for small and independent maintainers, and direct contact with the laboratory's security team. Last week, a critical vulnerability was disclosed on the BTCPay server, which attackers have exploited to empty merchants' Lightning nodes.