More than three dozen Bitcoin and crypto companies are calling on leading AI labs to provide open-source security researchers with early access to their most advanced models. They argue that the developers protecting hundreds of billions of dollars in digital infrastructure often have access to weaker AI tools than the attackers targeting those systems.
The letter, coordinated by the Bitcoin Policy Institute and published earlier this week, was signed by major industry names including Coinbase, Block, BitGo, Blockstream, Anchorage Digital, ARK Invest, Bitwise, Foundry, Casa and Exodus. Nonprofit developer organizations such as Brink, Chaincode and Btrust also backed the request.
At the heart of the complaint is the limited access available to Bitcoin Core developers, the relatively small group responsible for maintaining the software underpinning the Bitcoin network. According to the signatories, these developers generally cannot use the most powerful AI systems that labs make available to selected security partners.
Public AI models create another problem. Safety protections intended to prevent users from generating malware can also interfere with legitimate security research, making it harder for developers to identify vulnerabilities before attackers discover and exploit them.
As a result, researchers are often forced to rely on open-weight models that can be downloaded freely but typically offer less advanced capabilities.
The letter argues that attackers do not face comparable restrictions. AI companies and a limited number of trusted partners can gain access to emerging offensive capabilities months before the wider public, while similar tools eventually become available through public models, compromised corporate systems and specialized hacking software.
What Crypto Companies Want From AI Labs
The companies outlined five main requests: early access to the most capable cybersecurity-focused AI models, including before public launch; sufficient computing resources to conduct meaningful security assessments; protected environments for analyzing private code; access programs that include small teams and independent developers rather than only major organizations; and direct communication channels with AI labs’ security teams so researchers can report vulnerabilities and findings.
The timing of the appeal is significant. Two companies that signed the letter have recently experienced events that illustrate the security gap they are describing.
BTCPay Server, a signatory, disclosed a critical vulnerability last week that had already been exploited to drain Lightning nodes operated by merchants. Foundation, a hardware-wallet manufacturer that also backed the letter, lost one of its nodes during the same attack.
Following the incident, BTCPay highlighted how AI is changing the balance between cyber attackers and defenders by making it faster and less expensive to examine large software codebases for potential vulnerabilities.
In this case, however, the vulnerability was identified by defenders rather than attackers. A volunteer organization known as the Bitcoin Red Team began using AI models to analyze Bitcoin-related codebases earlier this month and has reported thousands of potential issues across hundreds of projects, including the finding that ultimately led to BTCPay’s security patch.
About The Author
More Stories
CFTC Urges Prediction Markets to Fix Incentive Filings to Drive Trading
Morgan Stanley-Backed Zerohash Denied Bid to Become a U.S. Trust Bank
Bitcoin Holds Near $63.5K as July PPI Comes in Below Expectations