Orion-1在网络攻击基准测试中击败GPT-6 Sol，攻击重建得分达80.8
artemis security推出orion-1，一款用于调查和响应网络攻击的ai模型，超越了claude opus 5.5和gpt-6 sol等通用ai模型。这对面临ai防护措施阻碍防御者的安全团队来说很重要。
artemis security launched orion-1, an ai model designed to investigate and respond to cyberattacks on corporate networks. orion-1 connects log data from networks, endpoints, and cloud services to trace attacks back to a single actor and initiate countermeasures. on artemis' benchmark, orion-1 outperformed general-purpose models like claude opus 5.5 and gpt-6 sol, scoring 80.8 in attack reconstruction compared to 58.3 for gpt-6 sol. orion-1 is live in private preview for select customers on the artemis platform. this launch addresses challenges faced by security teams as ai guardrails meant to block attackers are now blocking defenders.