Learn how to secure Model Context Protocol (MCP) deployments with granular policy enforcement and post-quantum cryptography for prompt engineering.
The post Granular Policy Enforcement for Quantum-Secure Prompt Engineering appeared first on Security Boulevard.
Most breaches don’t start with malware or zero-day exploits. They start with a login. An attacker gets hold of a password, maybe through phishing, reuse, or a leaked credential dump. They test it against a remote system. An SSH prompt appears. The credentials work. From there, everything unfolds quietly – privilege escalation, lateral movement, persistence. By the time anyone notices, the damage is already done. […]
The post Preventing Breaches – MFA on Remote Access to Linux, Unix, and Infrastructure Systems appeared first on 12Port.
The post Preventing Breaches – MFA on Remote Access to Linux, Unix, and Infrastructure Systems appeared first on Security Boulevard.
AttackIQ has released a new attack graph that emulates the behaviors of LokiLocker ransomware, a .NET based strain active since at least mid-August 2021. The malware combines defense evasion and impact techniques, including disabling Task Manager and Windows Firewall, as well as deleting Volume Shadow Copies to hinder detection and prevent restoration.
The post Emulating the Systematic LokiLocker Ransomware appeared first on AttackIQ.
The post Emulating the Systematic LokiLocker Ransomware appeared first on Security Boulevard.
In our opinion, Gartner’s 2026 research reflects this broader evolution. Identity has expanded beyond perimeter controls and point-in-time authentication to encompass verification of the human, contextual risk assessment, and automated trust decisions.
The post The 2026 CISO Mandate: Proactive, Passwordless, and Context-Aware Identity Assurance appeared first on Security Boulevard.
Compliance expectations across SMB markets are rising as supply chain regulations and cyber insurance requirements raise the baseline for security maturity. Regulatory standards such as CIS Controls v8, the NIS2 Directive, ISO 27001, SOC 2, PCI DSS, HIPAA, Cyber Essentials, CMMC 2.0, DORA, and the Essential Eight now shape what that baseline looks like.
The post MSP Strategic Defense: Building Compliance on Dynamic Attack Surface Reduction appeared first on Security Boulevard.
An analysis of more than two trillion IT events collected during 2025 by Barracuda Networks finds 90% of ransomware incidents exploited firewalls via unpatched software or a vulnerable account that enables cybercriminals to gain access to an IT environment. Merium Khalid, director of offensive security for the security operations center (SOC) at Barracuda Networks, said..
The post Analysis: Root Cause of Most Security Incidents Traced to Unpatched Firewalls appeared first on Security Boulevard.
OpenClaw runs locally, but the risk lives in SaaS. Learn how OAuth tokens, API access, and AI agents create identity-based exposure across platforms.
The post OpenClaw Security Risk: OAuth and SaaS Identity appeared first on Security Boulevard.
AI agents have moved from experimental to operational. Now security teams must verify agent identity, intent, and trust before automation becomes exploitation.
The post Early 2026 is Proving KasadaIQ’s AI Predictions appeared first on Security Boulevard.
When Microsoft confirmed that a bug allowed Copilot to surface and summarize emails marked confidential despite existing DLP controls, it reignited urgent questions about Microsoft Copilot security, DLP bypass risk and enterprise AI data protection. The reaction was immediate.
For many CISOs and security leaders responsible for Microsoft 365 security and AI risk management, it was a recognition of some exposure risks they were very much aware of.
AI assistants like Microsoft Copilot do not invent risk. They illuminate it. In this case, the Copilot DLP bypass highlighted a deeper enterprise AI security issue: data trust across Microsoft 365 and SaaS environments is not where it needs to be.
Copilot is easy to adopt. It integrates seamlessly into the Microsoft ecosystem. It promises productivity gains without heavy infrastructure changes. For organizations under pressure to move quickly into an AI-enabled future, enabling it feels natural and even necessary.
The controls were in place. The Microsoft 365 tenant was secured. DLP policies were configured.
Yet speed has a cost.
In the rush to capture AI value, many teams did not fully evaluate how sensitive data was classified, how permissions had expanded over time or how enforcement behaved under new AI-driven access patterns. What this moment exposed was not a single bug, but a broader condition: structural data fragility.
Over years, sensitive data sprawl, permission creep and incomplete classification quietly compound. The environment remains operational. Compliance boxes are checked. But structural risk accumulates beneath the surface, largely invisible until AI begins operating across it at scale.
How did the Microsoft Copilot DLP bypass reveal a data trust gap?AI operates at machine speed. It searches, correlates and summarizes across vast volumes of content in seconds. If sensitive information is accessible within the environment, it will surface.
The core issue is not simply that Copilot bypassed DLP controls. The deeper issue is that protection depends on trust in your data foundation. Trust that you know what is sensitive, that it is accurately classified and that policies are enforced consistently across the estate.
Without that trust, controls become fragile.
For years, unstructured data has multiplied across cloud drives, SaaS, collaboration platforms and email systems. Files were shared. Access was granted. Exceptions were made. Policies were written but rarely continuously validated.
AI did not create this complexity. It revealed it.
Are AI copilots exposing hidden DLP gaps and data exposure risks?This event may not be the last headline of its kind.
As organizations move quickly toward AI copilots, GenAI applications and agentic workflows, new edge cases will emerge. AI interacts with data differently than humans do. It traverses context. It aggregates at scale. It tests the boundaries of policy enforcement.
Unless data trust is embedded across the entire data estate, discovery, classification, detection, remediation, policy orchestration and prevention working in concert, similar exposures are a very real possibility for any company.
This is not a failure of innovation. It is a reminder that innovation amplifies whatever foundation it sits upon.
Is Microsoft ecosystem security enough without a strong data foundation?Many leaders assumed that operating inside a trusted vendor ecosystem meant they were inherently protected. Platform security is critical, but it cannot compensate for incomplete visibility into your own data landscape. AI exposes the difference between compliance and confidence.
Compliance asks, "Do we have DLP rules?"
Confidence states, "We truly understand what matters across our data estate."
Data trust means:
When these elements operate together, AI becomes an accelerator. When they do not, AI becomes a magnifier of existing gaps.
How can organizations build data trust and achieve DLP at AI speed?The Copilot story should not drive panic. It should drive reflection and action.
Organizations deserve to securely innovate at the speed of AI.
Data trust is not a feature, it's an architectural commitment. It requires moving beyond static rules and fragmented tools toward a unified approach that continuously understands, prioritizes and protects what matters most.
This is where modern, AI-native data security platforms like MIND play a critical role.
MIND was built to establish data trust at scale and deliver Stress-Free DLP. It continuously discovers sensitive data and enables on-demand classification across SaaS, on-premise file shares, endpoints, emails and emerging AI tools, so protection keeps pace with change. It applies context-aware detection, prioritizes real risk and automates remediation to reduce noise and remove manual toil. The result is DLP on Autopilot, an intelligent system that understands your business and protects what matters without slowing it down.
When data trust is embedded across your environment, AI initiatives are no longer a source of uncertainty and exposure risk. They become sustainable, measurable and defensible.
Security and AI innovation do not have to compete.
But without data trust, they will.
The future of enterprise AI security will belong to organizations that move fast, thoughtfully, with clarity about what matters and confidence in how Microsoft Copilot, DLP controls and data trust are continuously protected.
The post Microsoft Copilot DLP Bypass: A Data Trust Wake-Up Call for AI Security appeared first on Security Boulevard.
The post Life in the Swimlane with Chris Mallow, Partner Solutions Architect appeared first on AI Security Automation.
The post Life in the Swimlane with Chris Mallow, Partner Solutions Architect appeared first on Security Boulevard.
Session 13D: Software Security: Code and Compiler
Authors, Creators & Presenters: Ruishi Li (National University of Singapore), Bo Wang (National University of Singapore), Tianyu Li (National University of Singapore), Prateek Saxena (National University of Singapore), Ashish Kundu (Cisco Research)
PAPER
Translating C To Rust: Lessons From A User Study
Rust aims to offer full memory safety for programs, a guarantee that untamed C programs do not enjoy. How difficult is it to translate existing C code to Rust? To get a complementary view from that of automatic C to Rust translators, we report on a user study asking humans to translate real-world C programs to Rust. Our participants are able to produce safe Rust translations, whereas state-of-the-art automatic tools are not able to do so. Our analysis highlights that the high-level strategy taken by users departs significantly from those of automatic tools we study. We also find that users often choose zero-cost (static) abstractions for temporal safety, which addresses a predominant component of runtime costs in other full memory safety defenses. User-provided translations showcase a rich landscape of specialized strategies to translate the same C program in different ways to safe Rust, which future automatic translators can consider.
ABOUT NDSS
The Network and Distributed System Security Symposium (NDSS) fosters information exchange among researchers and practitioners of network and distributed system security. The target audience includes those interested in practical aspects of network and distributed system security, with a focus on actual system design and implementation. A major goal is to encourage and enable the Internet community to apply, deploy, and advance the state of available security technologies.
Our thanks to the Network and Distributed System Security (NDSS) Symposium for publishing their Creators, Authors and Presenter’s superb NDSS Symposium 2025 Conference content on the Organizations' YouTube Channel.
The post NDSS 2025 – Translating C To Rust: Lessons From A User Study appeared first on Security Boulevard.
Explore how AI-driven behavioral heuristics and post-quantum security protect Model Context Protocol (MCP) deployments from advanced AI-age threats.
The post AI-Driven Behavioral Heuristics for Quantum-Era Threat Detection appeared first on Security Boulevard.
The cybersecurity market was jolted last week after Anthropic dropped a bombshell announcement. The company’s new AI Claude model identified 500 previously unknown high-risk vulnerabilities hidden in widely used software. That is not a minor milestone. It is a technically significant achievement and a clear demonstration of how quickly AI capabilities are advancing.
What came next was just as dramatic as the markets responded almost immediately. Shares of several major cybersecurity firms plunged, with some posting double-digit declines within hours of the news. Investors reacted as though Anthropic’s breakthrough meant the displacement of commercial security tools and services as we know them.
The short answer is no. AI is not replacing cybersecurity vendors. It’s simply making them more capable.
AI is an extraordinarily capable tool. In this case, it was tasked with identifying vulnerabilities in software code and performed impressively. However, vulnerability discovery represents only one segment of a much broader and more complex cybersecurity ecosystem.
Cybersecurity vendors do far more than scan for weaknesses in code. They design and maintain extensive suite of interconnected product portfolios that integrate detection, prevention, response, analytics, orchestration, governance, and compliance capabilities. They operate at enterprise scale. They maintain deep relationships with enterprise clients, understand their operational constraints, and tune defenses against dynamic risks. They invest in research. They maintain global threat intelligence operations. They test and harden their solutions continuously. They work closely with customers to adapt to evolving business risks. And importantly, they incorporate disruptive innovation, such as AI, into their offerings. Their domain spans everything from threat detection and incident response to compliance, identity management, and data protection.
AI just delivered a technological wake-up call regarding vulnerability management. But most leading vendors have already been working to embed AI and ML into their products for some time.
While the stock plunge looked ominous, it’s actually signaling a healthy shift for the cybersecurity sector. Competition driven by valuable capabilities drive maturity. AI is accelerating innovation of efficiency and effectiveness.
Customers will have more freedom to choose the most capable providers rather than being bound by self-contained legacy platforms. Companies with the best ideas will earn market share through proven value, speed, and agility.
This leveling effect is what the cybersecurity industry needs. It encourages experimentation, accelerates technological maturity, and pushes all players to evolve. AI won’t destroy the cybersecurity market. It will make it stronger and more competitive.
The Anthropic announcement didn’t mark a collapse of the industry; it signaled its evolution. AI will empower developers to reduce vulnerabilities before release. It will empower attackers to identify and weaponize weaknesses faster. It will empower defenders to detect, patch, and validate more effectively. And it will empower emerging vendors to compete more directly with incumbents.
As AI flexes it powers the cybersecurity industry becomes more challenging and capable.
The cybersecurity market was jolted last week after Anthropic’s new AI Claude model identified 500 previously unknown high-risk vulnerabilities hidden in widely used software.
Shares of several major cybersecurity firms such as Palo Alto, Okta, and Crowdstrike cratered, with some posting double-digit losses!
Investors panicked in thinking this was upending cybersecurity vendors.
In reality, AI is not replacing such vendors, it will simply contribute to what they offer.
AI is a very powerful tool. It will be used by developers, attackers, and defenders to find vulnerabilities.
Anthropic’s announcement doesn’t mark the collapse of the cybersecurity industry. It is a signal of its evolution.
The post AI Shocks the Cybersecurity Market appeared first on Security Boulevard.
How Can Non-Human Identities Strengthen Cybersecurity? Are organizations truly leveraging the full potential of Non-Human Identities (NHIs) in their quest for robust cybersecurity? With cybersecurity threats continue to evolve, there’s a pressing need to adopt innovative solutions that go beyond traditional security measures. One such solution is the effective management of Non-Human Identities, especially in […]
The post Is the investment in Agentic AI justified by its cybersecurity benefits appeared first on Entro.
The post Is the investment in Agentic AI justified by its cybersecurity benefits appeared first on Security Boulevard.
The Role of Agentic AI in Modern Cybersecurity Are you exploring the future trends making waves in cybersecurity today? One transformative technology, Agentic AI, is reshaping how organizations manage Non-Human Identities (NHI) and Secrets Security Management, allowing businesses to stay ahead of potential cyber threats. Understanding Non-Human Identities Machine identities, also known as Non-Human Identities […]
The post How does Agentic AI help you stay ahead in cybersecurity appeared first on Entro.
The post How does Agentic AI help you stay ahead in cybersecurity appeared first on Security Boulevard.
What is Driving the Buzz Among Tech Leaders Over Non-Human Identities in Cybersecurity? Cybersecurity is evolving at an unprecedented pace, and amidst the dynamic shifts, Non-Human Identities (NHIs) have emerged as a pivotal component that tech leaders are increasingly excited about. The importance of NHIs extends beyond just another layer of security; it signals a […]
The post Why are tech leaders excited about NHI in cybersecurity appeared first on Entro.
The post Why are tech leaders excited about NHI in cybersecurity appeared first on Security Boulevard.
What Drives the Need for Non-Human Identity Management in Cloud-Native Security? How are organizations managing the security risks associated with non-human identities (NHIs) in their cloud environments? With digital transformation advances, the complexity and quantity of machine identities surpass human user identities. These NHIs, essentially machine identities, play an integral role in cloud-native security but […]
The post How does Agentic AI deliver value in cloud-native security appeared first on Entro.
The post How does Agentic AI deliver value in cloud-native security appeared first on Security Boulevard.
Session 13D: Software Security: Code and Compiler
Authors, Creators & Presenters: Nicolas Badoux (EPFL), Flavio Toffalini (Ruhr-Universität Bochum, EPFL), Yuseok Jeon (UNIST), Mathias Payer (EPFL)
PAPER
type++: Prohibiting Type Confusion with Inline Type Information
Type confusion, or bad casting, is a common C++ attack vector. Such vulnerabilities cause a program to interpret an object as belonging to a different type, enabling powerful attacks, like control-flow hijacking. C++ limits runtime checks to polymorphic classes because only those have inline type information. The lack of runtime type information throughout an object's lifetime makes it challenging to enforce continuous checks and thereby prevent type confusion during downcasting. Current solutions either record type information for all objects disjointly, incurring prohibitive runtime overhead, or restrict protection to a fraction of all objects. Our C++ dialect, type++, enforces the paradigm that each allocated object involved in downcasting carries type information throughout its lifetime, ensuring correctness by enabling type checks wherever and whenever necessary. As not just polymorphic objects but all objects are typed, all down-to casts can now be dynamically verified. Compared to existing solutions, our strategy greatly reduces runtime cost and enables type++ usage both during testing and as mitigation. Targeting SPEC CPU2006 and CPU2017, we compile and run 2,040 kLoC, while changing only 314 LoC. To help developers, our static analysis warns where code changes in target programs may be necessary. Running the compiled benchmarks results in negligible performance overhead (1.19% on SPEC CPU2006 and 0.82% on SPEC CPU2017) verifying a total of 90B casts (compared to 3.8B for the state-of-the-art, a 23× improvement). type++ discovers 122 type confusion issues in the SPEC CPU benchmarks among which 62 are new. Targeting Chromium, we change 229 LoC out of 35 MLoC to protect 94.6% of the classes that could be involved in downcasting vulnerabilities, while incurring only 0.98% runtime overhead compared to the baseline.
ABOUT NDSS
The Network and Distributed System Security Symposium (NDSS) fosters information exchange among researchers and practitioners of network and distributed system security. The target audience includes those interested in practical aspects of network and distributed system security, with a focus on actual system design and implementation. A major goal is to encourage and enable the Internet community to apply, deploy, and advance the state of available security technologies.
Our thanks to the Network and Distributed System Security (NDSS) Symposium for publishing their Creators, Authors and Presenter’s superb NDSS Symposium 2025 Conference content on the Organizations' YouTube Channel.
The post NDSS 2025 – type++: Prohibiting Type Confusion With Inline Type Information appeared first on Security Boulevard.
Exploitation of a maximum severity authentication bypass zero-day vulnerability affecting Cisco Catalyst SD-WAN Controller and Manager has been reported. Immediate patching is recommended to thwart ongoing attacks.
Key takeaways:On February 25, Cisco released a security advisory (cisco-sa-sdwan-rpa-EHchtZk) to address a maximum severity severity authentication bypass vulnerability in Cisco Catalyst SD-WAN Controller, formerly known as SD-WAN vSmart, and Cisco Catalyst SD-WAN Manager, formerly known as SD-WAN vManage.
CVE Description CVSSv3 CVE-2026-20127 Cisco Catalyst SD-WAN Controller/Manager Authentication Bypass Vulnerability 10.0 AnalysisCVE-2026-20127 is a critical severity authentication bypass vulnerability in Cisco’s Catalyst SD-WAN Controller and Cisco Catalyst SD-WAN Manager. A remote, unauthenticated attacker could exploit this vulnerability by sending crafted requests to an affected system, allowing them to log into an affected device as a high-privileged user. Using this access, the attacker could modify network configurations for the SD-WAN fabric. According to the advisory, this vulnerability has been exploited in the wild in limited attacks. The advisory further clarifies that this flaw affects vulnerable versions regardless of the device's configuration and no workaround steps are available, however temporary mitigation guidance is available in the security advisory.
CISA releases an Emergency Directive for CVE-2026-20127
Coinciding with the release of the security advisory for CVE-2026-20127, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) released emergency directive (ED) 26-03 titled Mitigate Vulnerabilities in Cisco SD-WAN Systems. The ED directs Federal Civilian Executive Branch (FCEB) agencies to take immediate action to identify any Cisco Software-Defined Wide-Area Networking (SD-WAN) systems. The ED notes that CVE-2026-20127 and CVE-2022-20775, a path traversal vulnerability affecting SD-WAN devices, pose imminent risk to federal networks. While the ED applies to FCEB agencies, any users who have not yet mitigated their SD-WAN devices for either of these CVEs should take immediate action as threat actors have been observed exploiting these vulnerabilities.
As ongoing exploitation has been observed, Cisco’s security advisory does include indicators of compromise which can aid defenders in identifying if their device has been compromised. Nation state-sponsored actors, including Salt Typhoon and Volt Typhoon have been known for past exploitation of Cisco devices, so it’s imperative that immediate action is taken to remediate these vulnerabilities.
In addition to CISA, the Australian Signals Directorate’s Australian Cyber Security Centre (ASD’s ACSC) also released an alert warning of exploitation of CVE-2026-20127. The ACSC was credited in the Cisco security advisory for reporting the flaw to Cisco and the ACSC alert also includes a threat hunting guide co-authored by multiple agencies including CISA, the National Security Agency (NSA), the Canadian Centre for Cyber Security (Cyber Centre), the New Zealand National Cyber Security Centre (NCSC-NZ) and the United Kingdom National Cyber Security Centre (NCSC-UK).
Exploitation attributed to UAT-8616
While the alerts from the government agencies and Cisco's security advisory did not provide attribution for the attacks targeting CVE-2026-20127, Cisco’s Talos threat intelligence team released a blog attributing the threat activity to UAT-8616. Cisco Talos notes that UAT-8616 is assessed “with high confidence” as “a highly sophisticated cyber threat actor.” The blog by Cisco Talos includes guidance for investigating compromised devices as well as details the exploitation activity that they have observed.
Proof of conceptAt the time this blog was published on February 25, no public proof-of-concept (PoC) exploit had been identified. We anticipate that if a PoC is released, additional attackers will begin to leverage the exploit to conduct mass scanning and exploitation against vulnerable devices.
SolutionCisco has released patches for affected versions of Cisco Catalyst SD-WAN devices as outlined in the table below:
Affected Version Fixed Version Versions prior to 20.9 Migrate to a fixed release 20.9 20.9.8.2 (Estimated to be released on February 27) 20.11 20.12.6.1 20.12.5 20.12.5.3 20.12.6 20.12.6.1 20.13 20.15.4.2 20.14 20.15.4.2 20.15 20.15.4.2 20.16 20.18.2.1 20.18 20.18.2.1The advisory notes that versions 20.11, 20.13, 20.14, 20.16 and versions prior to 20.9 have reached their end of maintenance and customers should upgrade to a supported release.
Identifying affected systemsA list of Tenable plugins for these vulnerabilities can be found on the individual CVE pages for CVE-2026-20127 and CVE-2022-20775 as they’re released. This link will display all available plugins for these vulnerabilities, including upcoming plugins in our Plugins Pipeline.
Additionally, customers can utilize Tenable Attack Surface Management to identify public facing assets running Cisco Catalyst SD-WAN devices by using the following query: Document Title contains Cisco Catalyst SD-WAN
Get more informationJoin Tenable's Research Special Operations (RSO) Team on Tenable Connect for further discussions on the latest cyber threats.
Learn more about Tenable One, the Exposure Management Platform for the modern attack surface.
The post CVE-2026-20127: Cisco Catalyst SD-WAN Controller/Manager Zero-Day Authentication Bypass Vulnerability Exploited in the Wild appeared first on Security Boulevard.
AttackIQ has released a new attack graph that emulates the behaviors exhibited by BlackByte ransomware, a strain operated under the Ransomware-as-a-Service (RaaS) model that emerged in July 2021. Since its emergence, BlackByte has targeted organizations worldwide, including entities within U.S. critical infrastructure sectors such as Government, Financial Services, Manufacturing, and Energy.
The post Emulating the Mutative BlackByte Ransomware appeared first on AttackIQ.
The post Emulating the Mutative BlackByte Ransomware appeared first on Security Boulevard.
