AI agents breach simulated corporate networks in Vals AI benchmark
Vals AI’s CyberOffense benchmark tests whether AI agents can independently compromise realistic simulated corporate networks, with every action externally logged and automatically scored. The Decoder reports that agents progressed from web applications to Windows servers and administrative privileges, while the harder Lighthouse test covered lateral movement into a second company. Results varied by test: the report presents conflicting Nova leader claims and score figures, while Lighthouse results reportedly favored GPT-6 Sol Daybreak over Deepseek V4.1 Flash 1017.
The benchmark shows that several models can perform multi-step network intrusion without human assistance, although the available report contains inconsistent Nova scores and leader claims.
The gap between open and closed AI models in cyber capabilities keeps shrinking
Matthias Bastian View the LinkedIn Profile of Matthias Bastian
Nov 7, 2026
Nano Banana Pro prompted by THE DECODER
AI agents can break into realistic corporate networks on their own, Vals AI finds
#Vals AI#CyberOffense benchmark#GPT-6 Sol Daybreak#Deepseek V4.1 Flash 1017#Kimi K3.1 Pro#Meta Wonder#Grok 4.7#Kimi K3#GLM 5.3 Flash#OpenAI#AI Security Institute#Anthropic