Cloud Security Alliance

Model Context Protocol is quickly becoming a common way for AI agents to connect to the systems around them. Through MCP, an agent can search a repository, query a database, create a ticket, send a message, update a record, invoke an API, or trigger a workflow. That makes interoperability much easier. It also changes the security significance of something that can look fairly harmless: the tool catalog. Consider an agent that can see four tools: searchDocuments() sendEmail() creat…

AI is expanding faster than anyone can imagine, as organizations race to use AI to gain a competitive advantage. It started with Generative AI. Now, agentic AI is changing the way people think and work. With the rise in usage of AI agents, the fear of data leakage and having more than necessary permissions has risen exponentially. Agents are everywhere and organizations still do not have proven solutions to put guardrails around them to ensure secure usage and to prevent d…

In June, Google DeepMind did something unusual: it published an AI control roadmap premised on the idea that its own AI agents might turn against it. The AI Control Roadmap treats advanced internal agents as potential insider threats — the way a company treats a rogue employee who already has badge access. It builds on MITRE ATT&CK with a new taxonomy, TRAIT&R (Taxonomy of Rogue AI Tactics and Routines), covering loss of control, work sabotage, and direct harm. And it sequences…

Research conducted in collaboration with Jazz finds that, among respondents experiencing data loss, only 16% said DLP blocked their most significant incidents before data left the organization  SEATTLE – Oct. 13, 2026 — Data loss prevention (DLP) tools are often functioning as after-the-fact forensic systems rather than prevention controls, according to DLP in the AI Era, a new survey from the Cloud Security Alliance (CSA), the world’s leading not-for-profit organization committed t…

  Why the industry needs The Builder Exchange Software development traditionally ran through a narrow channel that involved a defined set of engineers, pipelines, and a review process security could have control over. That model worked because building took experience and effort to execute. Now, AI app generators let anyone in an organization describe an app and ship it to production in an afternoon. The speed at which AI has enabled citizen developers, software builders who don’t…

  TL; DR Learn how transitioning to an AI-driven exposure management strategy enabled Tenable’s Chief Security Officer, Robert Huber, to conquer software sprawl, integrate fragmented data, manage the risks of rapid AI adoption, and pivot from sharing highly technical statistics to conveying corporate risk in a way that executives and board members can easily comprehend.   Key Takeaways A comprehensive exposure management program equips security leaders with the precise …

  Hybrid environments demand end-to-end visibility to stop modern attacks Hybrid environments are a dominant trend in enterprise technology, but they continue to present unique issues to the defenders tasked with securing them. By 2026, Gartner predicts that 75% of organizations will adopt hybrid cloud strategies [1]. At the same time, only 23% of organizations report full visibility across cloud environments [2]. That means a strong majority of organizations do not have comprehen…

  Metrics, logs, and traces pile up by the terabyte. Without the relationships between them, none of it adds up to an answer. Every cloud team has more observability data today than it did five years ago, by an order of magnitude. Metrics platforms, log aggregators, distributed tracing, APM agents on every service, synthetic checks running around the clock. And yet ask most SREs whether incidents get resolved faster than they used to, and the honest answer is usually “not really, …

TL;DR: OWASP’s 2026 rankings reflect the growing consequences of giving LLM applications access to tools and enterprise systems. Prompt injection remains No. 1, while excessive agency rises to No. 3 and several categories expand to cover newer deployment risks. Security teams should assume models will sometimes fail, then enforce identity-based, least-privilege access outside the model. It didn’t take long for large language models to move beyond the chat window. Today’s deployments incr…

  Review the code. Prove the agent is safe. Govern every call after — in your cluster. Artificial intelligence has entered a new phase. Organizations are no longer experimenting with simple chatbots that answer questions. Enterprises are deploying autonomous AI agents capable of reasoning, invoking tools, accessing enterprise systems, and making decisions that directly affect customers, operations, and revenue. This shift from conversational AI to agentic AI creates opportunity — …

Zero Trust implementations rarely fail because an organization forgot to buy a security tool. More often, progress stalls because identity, cloud, data, and network security teams pursue separate initiatives without a shared roadmap. Effective Zero Trust program management brings discipline to this complexity. It aligns strategic intent with operational execution, coordinates interdependent projects, and connects security investments to measurable business outcomes. A formally chart…

Walk through almost any factory and you will see large red emergency-stop buttons on dangerous equipment. They are deliberately obvious and accessible because, when machinery malfunctions, operators must be able to stop it immediately. Unmanned rockets have flight-termination systems to prevent an off-course vehicle from causing uncontrolled damage. Commercial fleets use speed governors, GPS tracking, and remote shutdown capabilities to address theft or dangerous driver behavior. These…

  Executive Summary Post-quantum cryptography (PQC) is often framed as a cryptographic upgrade. For enterprise DNS, however, it is something more consequential: a stress test of operational maturity. As organizations prepare for a changing cryptographic landscape, long-standing gaps in DNS ownership, visibility, configuration, and governance will become harder to ignore. The organizations that use PQC as a catalyst to modernize DNS operations will be better prepared for the next in…

In April 2026 the Cloud Security Alliance, working with our partners SANS and RSAC, first published The “AI Vulnerability Storm”: Building a “Mythos-ready” Security Program that opened the floodgates on our research for understanding and preparing for the emergence of advanced, adversarial AI. The goal is to determine how we, as defenders, can best prepare our organizations for the inevitable (and not too far off) day where attackers can easily discover Zero day vulnerabilities, weaponize…

A sparkle icon shows up in the CRM. Nobody in IT deployed it. Account executives start using it to summarize renewal risk across their book, and the help desk is fielding tickets about whether it’s allowed. Security can’t answer with evidence. This is now the normal state of enterprise SaaS. Zoom gave admins four days in July 2024 to opt out before AI Companion auto-enabled. Microsoft 365 Copilot switches on in the admin center for every admin once a tenant holds one paid license. Google…

On nearly all sales calls, we receive questions around how we’re using AI on the pen test team, how AI is going to change our processes, and how fast these changes will be implemented. And these are all fair questions given the increasing presence of AI in the broader penetration testing space lately. Models from Anthropic and OpenAI are starting to find real vulnerabilities in open-source tools on their own, and the autonomous testing vendors are getting louder. In this post, we’re goi…

Modern stadiums are infrastructure unlike any other. I’ve written before on event day sparking stadiums into life with shops and food stands, transport hubs, vast telecommunications infrastructure, field-side technology and beyond, acting as one super-sized, connected ecosystem. Stadiums’ scale and complexity make them some of the toughest environments in cybersecurity. Now, we’re adding AI to those operations and bringing a new dimension of risk. The benefits of AI in stadium operations…

AI agents select tools, invoke APIs, retrieve data, act under delegated user context, and create outbound connections at runtime. That autonomy can make them useful. It also means you need to know which agents can reach which tools, models, APIs, and data sources, and in what context. Zero Trust microsegmentation can help. By enforcing explicit, fine-grained communication controls, organizations can: Reduce ambient service reachability Deny unapproved egress Limit the damage a com…

  Executive Summary The $545,000 Surfside Beach fraud demonstrates a growing security blind spot: attackers don’t need to breach your network when they can simply impersonate your domain. By registering lookalike domains and exploiting trusted communications, threat actors can bypass traditional perimeter, identity, and email defenses without compromising a single account. Continuous DNS posture management helps close this gap by detecting typosquatting and domain impersonation ear…

As organizations prepare for post quantum cryptography, I am seeing confusion in two areas. The first is around the algorithms themselves. ML KEM, ML DSA, SLH DSA, FIPS 203, 204 and 205 are often discussed as if they are interchangeable. The second is more important. Teams are starting to talk about migration before they understand what data, systems and business services depend on the cryptography they are trying to replace. PQC is not simply an RSA replacement program, it’s a data se…