TLDR
- OpenAI will provide $1 billion in subsidized access to its cybersecurity model over the next six months.
- The support covers discounted Daybreak access, technical assistance, and cybersecurity training.
- About 2,000 organizations already use Daybreak, according to OpenAI.
- Priority users include utilities, governments, regional banks, nonprofits, and open-source software maintainers.
- Crypto firms are not directly named, but open-source blockchain projects could still qualify.
- OpenAI said Astra can detect unknown software flaws and turn them into working attacks with limited human guidance.
OpenAI will provide $1 billion in subsidized access to its cybersecurity models over the next six months. The program aims to help organizations defend against AI-enabled attacks and software flaws that attackers could exploit before developers release fixes.
The support covers discounted access to Daybreak, OpenAI’s cybersecurity product, along with training and technical help. OpenAI says about 2,000 organizations already use the service.
OpenAI Expands Daybreak Cybersecurity Program
Daybreak offers two service levels. Blue uses OpenAI’s standard models for routine defensive work. Red gives approved organizations access to specialized cyber models for more sensitive security tasks.
The company will focus on water systems, electricity providers, state and local governments, community banks, nonprofits, and open-source maintainers. Crypto exchanges, custodians, and blockchain networks do not appear on the priority list.
Open-source crypto projects may still fit the program because many rely on small development teams. Bitcoin Core, Ethereum clients, and DeFi libraries support systems that protect large amounts of digital assets.
The timing follows recent security problems across crypto software. A flaw in BTCPay Server exposed credentials tied to Lightning nodes in August. Attackers used the weakness to steal funds before developers issued a fix.
Core Lightning also advised operators to disconnect systems after AI-generated vulnerability reports revealed several real flaws. Coldcard later released new firmware after a major theft and said frontier AI models helped detect unrelated bugs.
Astra Raises New Security Concerns
OpenAI also disclosed that its Astra model can identify unknown software flaws and build working attacks without step-by-step human guidance. That ability has increased concern over how advanced AI tools could change cyber risk.
More than three dozen companies, including Coinbase, Block, BitGo, Blockstream, and ARK Invest, later signed an open letter to AI labs. They asked for earlier access to powerful models for defensive security work.
The companies said developers responsible for major crypto systems often receive weaker tools than attackers. Their request focused on giving security teams better access to advanced AI models before criminals can use similar systems.
Eligible organizations can apply through the Daybreak website. OpenAI also plans to expand the program to partner countries in the coming weeks. The rollout will test whether defensive access can help organizations respond to threats faster.
