OpenAI prepares security sprint and expands defense support
What happened
On November 3, OpenAI said it plans a next internal security sprint, working with newly added researchers and security partner Cathedral Systems to test and harden its systems using its own models. In the previous sprint, 25% of product engineers used Codex to investigate and reproduce vulnerabilities and prepare patches, the report said. It also said rewards for key findings in the security bug bounty program have risen more than fourfold. OpenAI pledged $1 billion in subsidized access over six months for frontline defenders protecting critical services and digital infrastructure. The next sprint is still in preparation.
Written by AI from the coverage · updated 2 hours ago
Coverage
Follow the reports to see the story from different sides.
- OpenAI NewsPreparing our next security sprint to harden OpenAI
OpenAI 计划开展下一轮内部安全冲刺，联合新加入的研究人员与安全合作伙伴 Cathedral Systems，用自家模型测试并加固系统。安全漏洞赏金计划对关键发现的奖励提升逾四倍，上一轮冲刺中 25% 的产品工程师借助 Codex 调查发现、复现漏洞并准备补丁。OpenAI 还承诺在六个月内为保护关键服务与数字基础设施的一线防御者提供 10 亿美元的补贴访问额度。
Heat over time
Not enough continuous observations to draw a trend yet.