The Bitcoin Policy Institute and a coalition of crypto companies urged major AI labs to give Bitcoin developers early access to frontier models. The group said stronger tools could help defenders identify vulnerabilities before attackers exploit open-source financial infrastructure.
The BPI published the letter on Monday and called for standing trusted-access programs for qualified defenders. It said Bitcoin Core developers and other maintainers often lack access to advanced cyber programs run by AI labs.
The letter also said public frontier systems can block some security work through built-in guardrails. As a result, developers may depend on less capable open-weight models when examining software that supports digital and financial systems.
The African Bitcoin Institute, Anchorage Digital, BitGo, Bitwise, Blockstream, Bull Bitcoin, MARA, Kraken, Ledger and Trezor joined the appeal. The group said Bitcoin alone secures more than $1 trillion in value.
The BPI said frontier AI can scan large codebases, detect possible weaknesses and speed up difficult technical work. Those capabilities can support defenders, but adversaries can also use them to improve attacks.
The institute said several open-source maintainers had reported sophisticated actors using advanced AI capabilities to sustain attacks. Some reports involved actors that maintainers believed could include foreign adversaries.
Can open-source defenders keep pace if advanced cyber models remain harder for them to access? The letter argued that dedicated access programs could give maintainers stronger tools as threats evolve. Crypto hacking activity also rose sharply in April 2026. Malicious actors stole more than $634 million from cryptocurrency platforms during the month.
That marked the highest monthly total since the Bybit hack. The February 2025 attack helped push industry losses to roughly $1.4 billion that month.
Also Read: Bitcoin Futures Shorts Ease as Riot Lands $9.1B AI Data Center Deal
Security researchers have also warned about faster AI-assisted vulnerability discovery. Immunefi CEO Mitchell Amador described the emerging threat as a “vulnerability apocalypse” as advanced models improve their ability to find software flaws.
OpenAI expanded its Daybreak program on August 10 with Daybreak Blue and Daybreak Red for approved cybersecurity defenders. It also introduced GPT-5.6-Cyber for vulnerability research, exploit validation and authorized security testing. OpenAI said GPT-5.6-Cyber completed 95% of requests in an internal advanced cybersecurity evaluation. The standard GPT-5.6 Sol model completed 1.5% of requests in the same evaluation.
Meanwhile, the industry has also increased funding for Bitcoin security. In July, Strategy, BlackRock, and Coinbase helped launch the Bitcoin Security Consortium. Members of the consortium pledged $15 million over three years to support Bitcoin developers and security researchers. The AI-access campaign targets similar defenders by seeking access to stronger cybersecurity tools.
The coalition does not seek unrestricted access to offensive-capable AI models. Instead, it wants smaller nonprofits, independent maintainers and open-source security teams to qualify for advanced defensive systems.
The crypto coalition wants trusted frontier AI access for Bitcoin and open-source defenders as cyber threats become more advanced. The request follows rising hacking losses, new AI security programs, and a separate $15 million effort to support Bitcoin developers and security researchers.