Why does a one hour nap during the day seem like a long nap but when you stay up all night and you have one hour to sleep, you think well an hour… might as well stay awake….

Episode 67 "Cyber Defense Matrix and Does Cloud Security Have to DIE to Win?" of Cloud Security Podcast where hosts Dr. Anton Chuvakin and Timothy Peacock interview Sounil Yu (Sounil Yu) about his Cyber Defense Matrix in the cloud and his new DIE model cloud.withgoogle.com/cloudsecurity/…

The march toward automated & autonomic security will continue. But, there are severe consequences of leaving it to mature on its own. Check out these tips from Sounil Yu on how to gently push our AI, ML, & automated technologies in the right direction. darkreading.com/operations/bui…

Give me thumbs up and retweet if LGBT is welcome and safe on your account 🏳️‍🌈 👍

@sirsquishy79 You might want to google me.

Help support Secure Controls Framework Council LLC by donating or retweeting. paypal.com/donate/?hosted…

A new favorite on my podcast list ! podcasts.apple.com/us/podcast/the…

I am speaking at Idaho Office of Emergency Management's Annual Preparedness and Cybersecurity Conference. Please check out my talk if you're attending the event! - via #Whova event app

George Takei

🚨 How a Prompt Injection can pass through Agentic AI Workflows undetected 🤖➡️🧠 Our new research shows how a single hidden instruction can silently flow through a multi-agent AI system – bypassing the interface chatbot, surviving summarization, and only activating when a

⚡𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗥𝗮𝗱𝗮𝗿 + 𝗢𝗽𝗲𝗻𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀 𝗦𝗗𝗞⚡ Our open-source scanner now supports OpenAI Agents SDK – helping developers visualize multi-agent workflows, track tool usage, and flag security risks with ease. We built an airline support demo with: ✈️ A

𝗛𝗲𝗮𝗱𝗶𝗻𝗴 𝘁𝗼 RSAConference 𝟮𝟬𝟮𝟱? 🌉🔐 Our team will be at 𝗥𝗦𝗔𝗖 𝟮𝟬𝟮𝟱, meeting with cybersecurity leaders and innovators to discuss how enterprises can safeguard AI Assistants and Agentic Workflows from development to deployment. If securing AI at scale is one

🎙️𝗥𝗲𝗱 𝘁𝗲𝗮𝗺𝗶𝗻𝗴 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀: 𝗪𝗵𝗲𝗿𝗲 𝗧𝗼 𝗘𝘃𝗲𝗻 𝗦𝘁𝗮𝗿𝘁? 🤔 SPLX founders joined the OWASP GenAI Security Project - Top 10 For LLM Apps GenAI Security Podcast with Aubrey King to dive into: 🧠 Continuous AI red teaming 🔁 How to build scalable testing workflows 🧩 Security risks in RAG,

Next week is our next run of our Attacking AI course! Check out the expanded syllabus ⬇️ payhip.com/b/xysOk 📢 Last Min Giveaway Time! Two seats up for grabs, winner will be chosen Tuesday next week! Each person can have up to 3 entries to the giveaway! ➡️Repost

📢 Arcanum Information Security Freebie and a giveaway! Today I’m pulling a section from our paid course, Hacking Your Career, and giving it away for free. ⬇️ It’s a crash course in skill-building with vetted training resources—organized by the three cybersecurity domains (red, blue,

🔎 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗥𝗮𝗱𝗮𝗿 𝗻𝗼𝘄 𝗱𝗲𝘁𝗲𝗰𝘁𝘀 𝗠𝗼𝗱𝗲𝗹 𝗖𝗼𝗻𝘁𝗲𝘅𝘁 𝗣𝗿𝗼𝘁𝗼𝗰𝗼𝗹 (Model Context Protocol (MCP)) 𝘀𝗲𝗿𝘃𝗲𝗿 𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗶𝗼𝗻𝘀 𝗶𝗻 𝗢𝗽𝗲𝗻𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗪𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀 – giving you visibility into how agents access external systems and helping you

Some notes from the floor: Most products do not leverage ML, just an LLM Most are using LLAMA hosted by the company. No they are not tuned or trained. Just system prompted. Many vendors replied data was local when it was obviously not Llms biggest killer feature?

⚠️ Multi-turn prompt injections are the next big threat to Agentic AI Workflows ⚠️ When targeting 𝗠𝘂𝗹𝘁𝗶-𝗔𝗴𝗲𝗻𝘁 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 (𝗠𝗔𝗦), prompt injection techniques become more complex – unfolding over multiple steps, priming agents across the conversation, bypassing

Check out my latest article: AI Threat Map v2.1 Navigating AI Threats & Tradeoffs for AI Adoption linkedin.com/pulse/ai-threa… via LinkedIn