The crypto industry demands that AI labs open access to advanced models: a question of survival
The Bitcoin Policy Institute and dozens of leading crypto industry companies have issued a collective appeal to the world's largest AI labs. The demand is simple and categorical: provide vetted open-source developers with reliable access to the most powerful artificial intelligence models. This is not about privilege, but about a vital measure to protect the financial infrastructure that underpins trillions of dollars.
The letter has been signed by dozens of organizations representing digital assets, cybersecurity, and open-source development. Among them are giants like Coinbase, Block, BitGo, and Trezor, as well as key players in the mining and infrastructure sectors, including MARA, Foundry, and Galaxy. All of them depend on the integrity of open infrastructure, wallets, and cryptographic libraries.
Why AI access has become a security issue
In recent weeks, it has become clear: digital asset security professionals are left without the most powerful tool in their arsenal. While attackers actively arm themselves with advanced AI models to find vulnerabilities and automate attacks, defenders are forced to work with less capable open-weight models. This creates a dangerous imbalance.
The scale of the threat is enormous. The Bitcoin network alone safeguards assets worth over a trillion dollars. When considering the entire ecosystem—from wallets and exchanges to payment networks and custodial systems—trillions of dollars are at risk. Behind this software are the personal savings of millions of people: pension funds, reserve funds, family wealth. A single unpatched vulnerability could wipe out an entire generation's fortune.
Defenders in the most vulnerable position
The situation is compounded by the fact that access to advanced models for legitimate security researchers is extremely limited and inconsistent. Many Bitcoin Core maintainers and other open-source developers have no access at all to the tools that attackers are already using. Public interfaces of the labs often block legitimate security testing work, forcing critical checks to be conducted on less powerful models.
Based on my information from open-source maintainers, the situation is even more serious: sophisticated attackers, likely operating from other countries, are conducting continuous AI-assisted attacks. Small teams find it extremely difficult to keep up with this pace. Even when systems remain protected, constantly reacting to complex activity diverts scarce resources away from development and routine security work.
What exactly is being demanded from the labs
The coalition proposes specific measures: creating or expanding permanent trusted access programs for qualified defenders of open financial infrastructure. This includes early, controlled access to the most powerful cyber models, and where appropriate, to pre-release versions. "Sufficient" computing power and the ability to run autonomous agents for long-running tasks are critical—without this, comprehensive security testing is impossible.
Special emphasis is placed on the need for secure environments for analyzing private or closed-source code, as well as the right of small non-profits and independent maintainers to participate on equal footing with large corporations and governments. Finally, a direct channel is needed for coordinated vulnerability reporting, remediation, and communication with lab security teams.
The digital asset industry is ready to help identify trusted participants, establish standards, and coordinate remediation efforts. Cutting-edge AI could become one of the most powerful defensive technologies in history, but only with fair access to such systems.
The letter was signed by representatives of the Bitcoin Policy Institute, Coinbase, Block, Blockstream, BitGo, Anchorage Digital, Bitwise, Trezor, MARA, Foundry, Galaxy, Strategy, Strike, and River, as well as dozens of other participants from the digital asset, cybersecurity, and open-source sectors.
My analysis: This appeal is not merely a request for access, but a signal of a fundamental shift in the balance of power. If AI labs ignore this call, we risk seeing a wave of incidents that will undermine trust in the entire crypto ecosystem. The question is not about technological advantage, but about the very survival of open financial infrastructure in an era of AI-armed attacks.