Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

datadog

LLM guardrails: Best practices for deploying LLM apps securely

Oct 22, 2025 By Tom Sobolik In Datadog
Prompt guardrails are a common first line of defense against client-level LLM application attacks, such as prompt injection and context poisoning. They’re also a critical component of a full defense-in-depth strategy for LLM security at the infrastructure, supply chain, and application level. The specific guardrails that teams implement depend highly on use case, but they are typically designed to.
Read Post
vanta

30+ due diligence questions to ask AI vendors in a security review

Oct 22, 2025 By Vanta In Vanta
Introducing third-party AI into your systems can be a milestone for productivity and growth, but it also expands your attack surface in unpredictable ways. If your AI vendors have weak controls, threats like data poisoning and algorithm failure can ripple through your systems.
Read Post
Torq

The Shift in SOC Escalation: From Manual to AI-Powered

Oct 22, 2025 By Patrick Orzechowski In Torq
Patrick Orzechowski (also known as “PO”) is Torq’s Field CISO, bringing his years of experience and expertise as a SOC leader to our customers. PO is a seasoned security veteran with a deep understanding of the modern security landscape. You can find him talking to SOC leaders and CISOs from major brands at cybersecurity events around the world. Every SOC has an escalation process — but not every SOC has one that is truly effective.
Read Post

The Smartest Way to Start a Cybersecurity Career | Marius Poskus

Oct 22, 2025 By Razorthorn Security In Razorthorn
Marius Poskus shares practical advice for those breaking into cybersecurity. He explains why beginners should start in Security Operations Centres (SOCs) rather than chasing rare pen‑testing roles. Building defensive skills first creates a stronger foundation and opens the door to any cybersecurity path—even in an AI‑driven future.
View Video
What Technologies Make Online Money Transfers Secure?

What Technologies Make Online Money Transfers Secure?

Oct 22, 2025 By SecuritySenses In SecuritySenses
A 2022 report by the Bank for International Settlements suggests that about $7.5 trillion is transferred daily around the globe. For context, the U.S. federal government spent $7.01 trillion in its 2025 fiscal year, which ran from October 2024 to September 2025, according to the U.S. Treasury Fiscal Data. Basically, this implies that about 7% more money is traded on the foreign exchange market daily than the U.S federal government spends annually.
Read Post
jfrog

CVE-2025-6515 Prompt Hijacking Attack - How Session Hijacking Affects MCP Ecosystems

Oct 21, 2025 By Ori Hollander In JFrog
JFrog Security Research recently discovered and disclosed multiple CVEs in oatpp-mcp – the Oat++ framework’s implementation of Anthropic’s Model Context Protocol (MCP) standard. Among these, CVE-2025-6515 stood out due to its potential threat of hijacking MCP session IDs. Within the context of MCP we’ve dubbed this new attack technique “Prompt Hijacking“. Your browser does not support the video tag.
Read Post

Are we only one prompt away from using AI for evil? #cybersecurity #ai #infosec

Oct 21, 2025 By LimaCharlie In LimaCharlie
Are we only one prompt away from using AI for evil? In this week's episode of The Cybersecurity Defenders Podcast, we explore a concerning reality about AI and cybersecurity. As AI becomes more prevalent within the threat actor community, exploits are being developed faster than humans can patch. The tools that help developers debug code can just as easily be used to weaponize vulnerabilities.
View Video
protecto

AI Privacy and Security: Key Risks & Protection Measures

Oct 21, 2025 By Anwita In Protecto
AI systems learn from vast amounts of data and then generalize. That power is useful and also risky. Sensitive data can slip into prompts. Proprietary datasets can be memorized by models. Attackers can steer models to reveal secrets or corrupt results. Meanwhile, your company is probably experimenting with multiple AI tools at once. That creates hidden data flows and inconsistent controls. “Traditional” app security isn’t enough.
Read Post

Unlocking AI's full value: CIO and CISO perspectives

Oct 21, 2025 By Tines In Tines
AI investment is accelerating across industries. Many organizations have either fully or extensively embedded AI in their business processes today. Yet, 40%* of IT teams still express a lack of trust in AI-generated outcomes. They remain in reactive mode, held back by disconnected systems, manual work, and rigid tools. What’s missing?
View Video

Copyright © 2026 OpsMatters™. All rights reserved.