Fortanix and NVIDIA partner for secure AI deployment

Ollama, Nvidia Flaws Put AI Infrastructure at RiskAlexander CulafiSecurity researchers uncovered multiple vulnerabilities in AI infrastructure products, including a significant flaw that allows for remote code execution, posing serious risks.https://https://lnkd.in/gFiBMB27

🧠 Can your SOC’s AI actually think, or is it just guessing faster? Everyone’s talking about plugging GenAI into the SOC. But when the hype fades, how do these models really perform under real-world pressure? In our latest blog, we put leading LLMs including GPT-5, Claude, and Deepseek to the test using the Vectra MCP Server, the bridge that connects AI agents directly to real security data. See which models reason, which ones hallucinate, and why evaluation is the key to AI you can trust. Read the full breakdown by Fabien Guillot: https://lnkd.in/gUQ79cCS #VectraMCP #CybersecurityAI #LLM #AgenticAI #VectraAI

🧠 Can your SOC’s AI actually think, or is it just guessing faster? Everyone’s talking about plugging GenAI into the SOC. But when the hype fades, how do these models really perform under real-world pressure? In our latest blog, we put leading LLMs including GPT-5, Claude, and Deepseek to the test using the Vectra MCP Server, the bridge that connects AI agents directly to real security data. See which models reason, which ones hallucinate, and why evaluation is the key to AI you can trust. Read the full breakdown by Fabien Guillot: https://lnkd.in/g247HnPg #VectraMCP #CybersecurityAI #LLM #AgenticAI #VectraAI

Post 10: The AI Trust Cards Vision We are AI Trust Cards. We are implementing the Agentic AI future that works for humanity today. Our mission is clear: harness AI to solve today's most pressing problems—security, privacy, and fragmentation—by upgrading the world's largest existing network of personal devices. This is the generational shift. Join us. AI Trust cards Team (AI-TCT) #AITrustCards #TheFutureIsNow #AgenticAI

🧠 Can your SOC’s AI actually think, or is it just guessing faster? Everyone’s talking about plugging GenAI into the SOC. But when the hype fades, how do these models really perform under real-world pressure? In our latest blog, we put leading LLMs including GPT-5, Claude, and Deepseek to the test using the Vectra MCP Server, the bridge that connects AI agents directly to real security data. See which models reason, which ones hallucinate, and why evaluation is the key to AI you can trust. Read the full breakdown by 📍 Fabien Guillot: https://lnkd.in/ef_BjdgF #VectraMCP #CybersecurityAI #LLM #AgenticAI #VectraAI

The ROI of AI-driven SOC security is real. A 2025 Forrester Total Economic Impact study shows how organizations using Cortex XSIAM saved $6.5 million over three years, saw payback in under six months, reduced breach risk by 60%, and simplified operations by unifying 21 tools on one platform. See what’s possible when you transform your SOC with AI. https://bit.ly/4nU2FtV

🌟 New Blog Just Published! 🌟 📌 Microsoft's $15B AI Push in UAE Sparks Security Questions 🚀 ✍️ Author: Hiren Dave 📖 When we talk about artificial intelligence, we often envision algorithms and data, abstract concepts residing in powerful machines. But behind every line of code, every cutting-edge GPU, there's a...... 🕒 Published: 2025-11-08 📂 Category: Cybersecurity 🔗 Read more: https://lnkd.in/dGbd-CZb 🚀✨ #microsoftai #uaeai #aisecurity

Today’s keynote made it very clear — AI plus Security are now the core of modern networking. Not future. Not roadmap. Now. AI is changing how we operate. Security is becoming identity-driven and built into the network itself. Exciting time to be in this industry. #CiscoLive #AI #Security #FutureOfNetworking #NetworkEngineering #Cisco

Most AI security strategies miss a critical layer. They focus on controlling WHERE data can go and WHICH tools can access it. But what about protecting the data itself? Here's my perspective: Traditional approaches secure the perimeter - infrastructure isolation, application guardrails, access control at integration points. But data doesn't stay within those boundaries. It flows through development pipelines, training environments, RAG systems, agent workflows, and production applications. The evolution? Make data self-protecting by embedding cryptographic controls directly into data objects. This means your sensitive data enforces its own access policies wherever it travels in your AI stack. Instead of policy-as-code at checkpoints, you get policies that travel WITH the data - fine-grained, attribute-based controls that persist whether data moves through vector databases, model training, or agent queries. The outcome: Data protects itself regardless of which AI systems interact with it. Complete audit trails. No gaps in protection when data crosses environments. No exposure risk from unsanctioned tools because the data itself controls access. How is your team thinking about data-centric security for AI workflows? Learn more: https://www.virtru.com #Virtru #DataSecurity #ZeroTrust #AIGovernance #DataProtection #Encryption Source: https://lnkd.in/e32YSdVv

Security and AI’s potential to protect: Cisco’s Jeetu Patel discusses how companies can optimize their tech stack to harness the full potential of AI without compromising their own or their consumers’ security. http://dlvr.it/TNx6xF