Skip to main content
Category

large language models

Cisco Report: No Frontier AI Model Is Immune to Attacks Cisco report finds no closed frontier AI model is safe from multi-turn attacks
adversarial AIAIAi benchmarksAI complianceAI governanceAI procurementAI safetyAI securityAmazonAnthropicattack success rateCiscoCisco AI DefenseCisco AI Threat ResearchCisco LLM Security LeaderboardCisco SystemsClaudeDeath by a Thousand PromptsEU AI ActFrontier ModelsGemini 3 ProgoogleGPT-5GPT-5.4GrokGrok 4.1 Fastlarge language modelsLLM securityMistralmodel cardsmulti-turn attacksNewsNIST AI RMFNIST Cyber AI ProfileNovaOpenAIred teamingThe-LatestxAI

Cisco Report: No Frontier AI Model Is Immune to Attacks

Bugcrowd Launches AI Training Environments for Software Vulnerability Detection Bugcrowd launches reinforcement learning environments to train AI on real software vulnerabilities
agentic AIAI agentsAI securityBlackbird VenturesBugcrowdCarnegie Mellon UniversityCostanoa VenturescybersecurityDARPA Cyber Grand ChallengeDave GerryDavid BrumleyExploitBenchfrontier AIfuzzingGeneral CatalystIndustry Ventureslarge language modelsMayhem SecurityNewsopen-source softwarePaladin Capital GroupRally Venturesreinforcement learningreinforcement learning environmentsSalesforce VenturesSecuritysymbolic executionThanassis AvgerinosThe-LatestTriangle Peak Partnersvulnerability research

Bugcrowd Launches AI Training Environments for Software Vulnerability Detection