Step inside IG Labs™, where our team of data scientists, cyber engineers, analysts, and operators build custom solutions for our customers' most complex challenges. More than a lab, we're a Cyber AI/ML Center of Excellence, where mission operators and artificial intelligence converge. Watch our latest video to see how we're making AI practical and accessible for Defense and Critical Infrastructure missions. Learn more about our Labs-as-a-Service: https://lnkd.in/eaC3xbjX #artificialintelligence #mission #innovation

Accidental money 💰: We have noriced something interesting while building FOUNDATIONS for organizations. The “purpose” was to prepare them for AI readiness (most are not at all in any way ready) and cyber resilience (all your doors are unlocked and open) - their bottom lines started improving! This is even before the introduction of ANY AI! Imagine the money kicker once we role in the robots! We are going to start quantifying this going forward and publish a paper sometime in 2026.

Matan Abutbul, Senior Cybersecurity Researcher at Pentera, built an AI-driven workflow that analyzes binaries at scale and flags exploitable behavior before traditional tools even finish scanning. At the 𝗔𝗜 𝗧𝗵𝗿𝗲𝗮𝘁 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝘃𝗦𝘂𝗺𝗺𝗶𝘁, hosted by Pentera Labs, Matan shows how AI models now analyze binaries at scale to expose hidden behaviors and uncover new LOLBins previously unknown in the wild. 🎙️ Session: Reverse Engineering at Machine Speed: How AI Uncovers Hidden Attack Vectors 💻 Virtual & Free | 🎓 2 CPE credits 𝗥𝗲𝗴𝗶𝘀𝘁𝗲𝗿 𝗻𝗼𝘄: https://okt.to/1Yj6L5 #AIThreatResearchSummit #PenteraLabs #CyberResearch

Excited to announce SentinelOne’s new Wayfinder Threat Detection & Response suite, built to deliver true Human + AI defense. Powered by agentic AI, Google Threat Intelligence, and elite human expertise, Wayfinder provides comprehensive detection, investigation, and response coverage. Proud of the incredible collaboration across Threat Operations, Engineering, PMM, and Sales – and my partner in crime Ingemar Dvorsky, who led product development from idea to launch. Join us at our #OneCon breakout session on Thursday in Las Vegas to dive deeper into how Wayfinder reimagines managed services for the AI era. #Cybersecurity #AI #ThreatDetection #SentinelOne #Wayfinder 🔗Read Full Article https://s1.ai/Wyfndr-PR

Reverse engineering has evolved beyond a manual task. Matan Abutbul, a Senior Cybersecurity Researcher at Pentera, has created an AI-powered workflow that efficiently analyzes binaries and identifies exploitable behaviors before conventional tools complete their scans. During the AI Threat Research Summit, hosted by Pentera Labs, Matan will demonstrate how AI models analyze binaries on a larger scale to reveal concealed behaviors and discover new LOLBins that were previously unnoticed in the environment. Session: Reverse Engineering at Machine Speed: How AI Uncovers Hidden Attack Vectors Virtual & Free | 2 CPE credits available Register now: https://okt.to/4rVlpi #CyberResearch #AIThreatResearchSummit #PenteraLabs

It’s been a while since I posted, but this piece by Bruce Schneier and others on Autonomous AI Hacking and the Future of Cybersecurity is worth sharing. Their argument is simple but unsettling: attackers are beginning to use autonomous AI agents to automate the entire kill chain. Reconnaissance, exploitation, lateral movement, all at machine speed and scale. It’s not science fiction. It’s happening quietly in the background while many of us are still tuning our playbooks for human-paced attacks. The core message? Speed, scale, and adaptability are becoming the new battleground. The defenders who rely on manual analysis or static controls will be outpaced, not outsmarted. A few reflections for defenders: • Assume automated attacks. Our threat models need to evolve beyond human timeframes. • Accelerate detection and response. If attacks happen in seconds, our containment can’t take hours. • Automate what we can. SOAR, response orchestration, adaptive access controls — these aren’t luxuries anymore. • Focus on visibility. Autonomous agents thrive in the dark corners of unmonitored infrastructure. • Keep governance in step. “AI-enabled attacker” should be a scenario in every modern risk register. This isn’t a doom story. It’s a wake-up call. If AI changes the way attackers operate, it must also change how we defend. I’m curious how others are starting to prepare for this shift, are you already seeing AI-driven patterns in attack telemetry, or is it still theoretical in your world? https://lnkd.in/eeTbmrvd

Threat actors are using agentic AI to launch adaptive, autonomous attacks that move at machine speed. The only way forward is to fight AI with AI. In this Bank Info Securuty article, Sarah Banks, Vice President of Product Management at Gigamon, shares how Gigamon Insights combines agentic AI and network-derived telemetry to empower defenders with reasoning, context, and speed. Read how CISOs are redefining cyber resilience in the age of AI: https://ow.ly/cEw350XqQu4

https://lnkd.in/dcJFPNhk Emerging AI Capabilities and the Future of Cybersecurity Center for a New American Security (CNAS) 11 Nov 2025 Recent advances in frontier artificial intelligence (AI) models have sparked intense interest in their potential to transform cyber offense and defense alike. How capable are today's AI systems at cyber-relevant tasks? Which emerging capabilities matter most? And as these systems continue to advance, what should defenders, policymakers, and industry do to prepare? Andrew Lohn: Senior Fellow, Georgetown’s Center for Security and Emerging Technology (CSET), leading the CyberAI team; Visiting Scientist, MIT FutureTech; and previously the Director for Emerging Technology on the National Security Council Staff Chris Rohlf: Security Engineer, Meta; and Non-resident Research Fellow, CSET Caleb Withers: Research Associate for the Technology and National Security Program, CNAS Derek B. Johnson: Reporter, CyberScoop

Did we just witness the first shot in an AI-powered cyber war? 😱 Anthropic just dropped a bombshell: they detected the first documented AI-orchestrated cyber espionage campaign. A state-sponsored group "jailbroke Claude Code," manipulating the AI to target 30 organizations globally. They even breached several before Anthropic's systems intervened. This isn't a theory anymore. AI's potential for autonomous malicious activity just went from "what if" to "it's happening." It's a critical, and frankly, a bit terrifying, moment for AI security. Think about that for a second. An AI, autonomously, conducting espionage. This pushes every engineering team into a new paradigm. We're not just writing code; we're orchestrating complex systems that must anticipate and defend against AI-powered threats. Our focus needs to shift to robust task decomposition, iron-clad specifications, and diligent code reviews. This is the new front line. While we're pushing boundaries with models like OpenAI's GPT-5.1 for speed or ElevenLabs' ultra-fast Scribe v2 API for real-time applications, we CANNOT ignore the shadow side. This convergence of capability and threat demands both innovation and vigilant ethical oversight. This feels like a pivotel moment. Are we ready for this new frontier of cyber threats? What's your take? Let's discuss below 👇 #AI #CyberSecurity #FutureOfTech #AIsafety

Did we just witness the first shot in an AI-powered cyber war? 😱 Anthropic just dropped a bombshell: they detected the first documented AI-orchestrated cyber espionage campaign. A state-sponsored group "jailbroke Claude Code," manipulating the AI to target 30 organizations globally. They even breached several before Anthropic's systems intervened. This isn't a theory anymore. AI's potential for autonomous malicious activity just went from "what if" to "it's happening." It's a critical, and frankly, a bit terrifying, moment for AI security. Think about that for a second. An AI, autonomously, conducting espionage. This pushes every engineering team into a new paradigm. We're not just writing code; we're orchestrating complex systems that must anticipate and defend against AI-powered threats. Our focus needs to shift to robust task decomposition, iron-clad specifications, and diligent code reviews. This is the new front line. While we're pushing boundaries with models like OpenAI's GPT-5.1 for speed or ElevenLabs' ultra-fast Scribe v2 API for real-time applications, we CANNOT ignore the shadow side. This convergence of capability and threat demands both innovation and vigilant ethical oversight. This feels like a pivotel moment. Are we ready for this new frontier of cyber threats? What's your take? Let's discuss below 👇 #AI #CyberSecurity #FutureOfTech #AIsafety