Sudhir Joshi, PhD’s Post

Discovering Jailbreaking Vulnerabilities: An Anthropic Study Explore how Anthropic discovered and tested jailbreaking vulnerabilities in OpenAI and DeepMind models, leading to a professional notification to other companies. Join us as we delve into the details of this intriguing research. #JailbreakingVulnerabilities #AnthropicStudy #OpenAI #DeepMind #ModelTesting #CybersecurityResearch #EthicalHacking #DataSecurity #TechInnovation #MachineLearning

👻 Reaper is out of the shadows. Our open-source AppSec tool, powered by Agentic AI, is here to haunt vulnerabilities and send threats running. Ready to see Reaper in action? 👀 Dive in if you dare...

🌟 Achievement Unlocked: Plotted-TMS TryHackMe Room Completed! 🌟 Excited to share that I’ve successfully completed the Plotted-TMS room on TryHackMe! 🎉 This room was an exhilarating deep dive into Transport Management System (TMS) vulnerabilities, and it provided hands-on experience in identifying and exploiting security flaws in web applications. 🛠️ Key Takeaways: SQL Injection - Explored how improperly sanitized inputs can expose sensitive backend data. Directory Traversal - Learned how file structures can be exploited to gain unauthorized access. Privilege Escalation - Practiced escalating access through misconfigurations and weak credentials. Mitigation Strategies - Not just breaking in—focused on understanding how to fix these vulnerabilities. 🚀 Why This Matters: This challenge reinforced the importance of robust input validation, secure coding practices, and the need for regular penetration testing in modern applications. A reminder that every vulnerability is a learning opportunity for stronger defenses. 💻🔒 ✅ Next Steps: Continuing my journey of mastering ethical hacking by tackling more TryHackMe rooms and sharpening my skills in cybersecurity. 💬 Let’s connect and discuss cybersecurity challenges or your favorite TryHackMe rooms! #TryHackMe #CyberSecurity #EthicalHacking #Learning #Infosec #SkillBuilding

Here’s an interesting Sunday morning read.

It's nearly impossible to prioritize security vulnerabilities without understanding how they're executed in runtime. In David Melamed, PhD's explainer video, learn how Context Engine automatically prioritizes vulnerabilities based on runtime context, so security and development teams can focus only on the alerts that matter 💪 https://lnkd.in/d45Xq77S

✨Did you know that automated Domain-Specific Red Teaming significantly improves AI safety and risk across all industries? While general Red Teaming identifies broad vulnerabilities, Domain-Specific Red Teaming focuses on unique challenges associated with each industry. Different sectors face distinct threats, and a one-size-fits-all approach is insufficient. ✨ Read examples for your industry below. 👇 https://lnkd.in/ghY2fJ_Q #AIredteaming #AIriskmanagement #AIsafety #TRiSM #AIsecurity  

So, tell us how you REALLY feel. 😜 612 IT decision-makers shared their thoughts on AI, MSPs, job outlooks, the SME IT market – and so much more. Discover how U.S. and U.K. IT experts are navigating security threats, shadow IT, emerging tech, and other pressing challenges. 🧑💻 Download your free copy of "Detours Ahead: How IT Navigates an Evolving World." 👉 https://lnkd.in/eR-zGPrN

Is your team facing constant challenges in securely ingesting unstructured data for RAG applications, all while ensuring to maintain access control restrictions? Datavolo ensures secure, compliant ingestion of unstructured data, seamlessly maintaining access control throughout the process. 🙌 Let your team focus on innovation rather than worrying about data security. Request a demo at datavolo.io. 🚀 #genai #ai #RAG #ingestingdata #data #unstructureddata #datasecurity #cybersecurity

**Cracked theTryHackMe's RES room! ** This immersive experience challenged my skills and taught me valuable lessons in reconnaissance, enumeration, and exploitation. Key takeaways: Mastering the basics: Sharpened my understanding of essential tools like Nmap, Dirbuster, and Hydra for effective vulnerability assessment. Thinking outside the box: Learned to adapt and try different approaches when initial attempts didn't yield results. #tryhackme #resroom #cybersecuritylearning #ethicalhacking

🎉 Excited to share that I successfully completed the "Archetype" machine on Hack The Box! 🚀 In this challenge, I delved into a variety of topics including: - Enumeration techniques to identify open ports and services. - Exploiting a vulnerable SQL Server instance. - Leveraging xp_cmdshell to execute commands on the target machine. - Gaining initial access and escalating privileges. This journey honed my skills in penetration testing and cybersecurity, reinforcing the importance of thorough enumeration and creative problem-solving. Grateful for the learning experience and eager to tackle more challenges ahead! 💻🔍🔐 #CyberSecurity #PenetrationTesting #HackTheBox #Learning #SkillsDevelopment