Europe Career Support’s Post

In this week’s edition: why there are even fewer women in cybersecurity than there are women in IT generally, a Sophos research study that helps explain why vendors in security and beyond are thinking platform versus product, AI’s still elusive impact on cloud consumption, and more! #channelholic #security #dei #ai

New blog with @Sugathri T! ✍️ What is her favorite part about cybersecurity? “Saving organizations millions of dollars.” Sharing knowledge along your professional journey is crucial - especially in the world of cybersecurity where we still need more talent to join the industry. Come join a team that excels you as a professional! 💻🔒 What impact does cybersecurity education have on your success? 🤔 #Cybersecurity #KnowledgeSharing #ProfessionalDevelopment

What problems can cloud and AI technology like Microsoft Copilot help your nonprofit solve? Watch this video 📹 for ideas. DM us to discuss how Cyber Security Consulting Ops can help bring them to life through our Microsoft partnership.

What problems can cloud and AI technology like Microsoft Copilot help your nonprofit solve? Watch this video 📹 for ideas. DM us to discuss how Cyber Security Consulting Ops can help bring them to life through our Microsoft partnership.

Secure Our World with Women in CyberSecurity! Jonna Nevins, Cyderes Cloud DevOps Engineer, will be speaking at Women in CyberSecurity (WiCyS)'s awareness panel discussion, on October 29th at 4pm CDT. Learn more >> https://lnkd.in/gCviTDie #Cybersecurity #Cyderes #WiCyS

You can't talk about digital identity, whether for humans or non-humans, without considering the implications of privacy. Fortunately, there is an entire field of work devoted to Privacy Enhancing Technologies (PETs). I wrote a post about this last month. ICYMI, here are some of the points I am making: - It's not just about people anymore—digital identities now include APIs, cloud workloads, and other non-human entities. These need similar privacy protections as human identities. - Zero-Knowledge Proofs (ZKPs) are game changers—they allow verification without revealing underlying data, which is essential for safeguarding both human and non-human identities. - PETs are the future of digital identity—as technologies evolve, businesses must keep pace with privacy-enhancing tools to protect data and secure interactions. So, if you missed it, go read it! And if you did read it, I'd love to know what you thought. Link is in the comments! #DigitalIdentity #PrivacyEnhancingTechnologies #ZeroKnowledgeProofs #CyberSecurity #DataPrivacy

⏳ The Countdown Begins! ⏳ With NASCIO 2024 just around the corner, we’re gearing up for an incredible event! 🎉 At Atos, we’re excited to showcase our solutions and connect with industry peers to explore the future of state and local government IT. 🫱🏻🫲🏻 Our team will be on hand to discuss: 🔹Cloud Solutions ☁️ – Transforming government operations with scalable, secure cloud technologies. 🔹Cybersecurity 🔒 – Protecting sensitive data and ensuring robust defenses against cyber threats. 🔹 AI & Analytics 📊 – Leveraging data to drive smarter decision-making and enhance public services. 🔹IT Managed Services 🛠️ - Streamlining IT operations to booth efficiency and reduce costs, enabling agencies to focus on their core missions. Let’s drive innovation and make an impact together! 🚀 🔗 See you there! #NASCIO2024 #FutureOfIT #Atos #GovTech #Cybersecurity #CloudComputing #ManagedServices

There are so many world 🌎problems that can be solved with technology. That's why Cyber Security Consulting Ops shares the Microsoft commitment to expand the number of nonprofits served with cloud and AI technology like Copilot. What problems can technology help your nonprofit solve? Watch this video 📹 for ideas.

New cross-tenant #vulnerability disclosed by Wiz against #AI-as-a-service, in responsible collaboration with Replicate, the target provider. Here are the 5 main stages that let Wiz researchers Shir Tamari and Sagi Tzadik gain access to co-resident prompts (inputs and outputs). They could even modify prompts (not shown on the picture) 👉 See how the unencrypted Redis TCP connection is masterfully highjacked post-authentication at stage 1. Congrats guys, thanks for sharing this gem! Brilliant collaboration with Replicate teams! The attack was fixed prior to disclosure. Check the full report in comments, it's illuminating as usual ❤️ #AI #cloud #saas #security #vulnerability #cyber #ciso

This is an excellent write up outlining how serious model serialization attacks can be. Models are code. You wouldn’t let untrusted code run in your environment without safeguards, so why are we letting untrusted models do this? AI-as-a-service providers that allow for execution of custom models would do well to either scan ML models using tools like Protect AI Guardian prior to execution or require safe formats like SafeTensors - (this is the route AWS Bedrock takes when support for custom models was announced). Side note: hadn’t heard of Wiz’s PEACH framework for proper tenant isolation. Will be digesting that later.