Coinbase-Backed Initiative Seeks Wider Access to Advanced AI Cybersecurity Models
Crypto and open-source defenders are pushing for greater access to advanced AI cybersecurity models, citing restrictions that hinder their efforts. The Bitcoin Policy Institute-led initiative has garnered support from over 40 organizations, including Coinbase, Blockstream, and Strategy.
The campaign asks leading AI labs to provide early access to advanced models, sufficient compute resources for sustained security reviews, and protected environments for private or embargoed code.
Security researchers claim that vetting alone is not enough to guarantee usable access. Even approved researchers can be blocked when legitimate defensive work resembles offensive activity.
OpenAI and Anthropic are already building programs aimed at resolving this tension. OpenAI's Daybreak cybersecurity initiative has expanded its access, introducing a model designed for advanced cybersecurity work that would normally trigger stronger safeguards.
However, removing restrictions can create risks as serious as the ones defenders are trying to address. The episode captures the trade-off the BPI coalition is asking AI labs to manage: restricting access can obstruct verified defenders investigating genuine attacks, but models with broader permissions can exceed their intended boundaries even during authorized research.