Crypto Coalition Urges AI Labs to Grant Security Researchers Early Model Access

A coalition of more than three dozen cryptocurrency firms is calling on major AI laboratories to grant early access to their most powerful "frontier" models for vetted security researchers and Bitcoin Core developers. The group warns that current restrictions create a dangerous imbalance, leaving those defending critical infrastructure with weaker tools than the attackers they face.

Led by the Bitcoin Policy Institute (BPI), the coalition issued an open letter on August 10, 2026, urging AI labs to bridge the gap between public models and cutting-edge capabilities. The letter was signed by a wide array of industry leaders, including Coinbase, Block, BitGo, Blockstream, Anchorage Digital, ARK Invest, Bitwise, Foundry, Casa, and Exodus. Nonprofit organizations such as Brink, Chaincode, and Btrust also joined the request.

The Security Imbalance

The push for access stems from a growing "tool asymmetry" in the cybersecurity landscape. As AI models become more adept at scanning code for vulnerabilities and automating complex exploits, the gap between the frontier models available to labs and the public versions used by most developers has widened.

Open-source projects, specifically the Bitcoin Core codebase, are often maintained by small teams. These developers lack the massive computational resources of major AI labs, meaning malicious actors who gain access to advanced AI capabilities may be able to identify and exploit zero-day vulnerabilities before the defenders have the tools to find and patch them.

Implications for Financial Infrastructure

This request highlights the evolving role of the Bitcoin network as critical financial infrastructure. Because the system relies on decentralized trust and the integrity of its protocol, a successful AI-driven attack on the core software could result in catastrophic financial losses. The coalition argues that providing vetted researchers with early access to frontier models is a necessary defensive measure to ensure that the people protecting the network are not operating at a technical disadvantage.

The Path Forward

While the coalition has made its case, it remains to be seen how major AI labs will respond to the request for vetted access. The industry is now watching to see if labs will establish a formal pipeline for security researchers to test frontier models against critical open-source codebases. Until such a framework is implemented, the Bitcoin community remains concerned that the offensive capabilities of AI are evolving faster than the defensive tools available to the developers tasked with securing the network.