Audio Prompt Injection: What Voice Agents Hear That Text Guardrails Miss

Key takeaways Only 24 percent of organizations can fully inspect AI traffic without impacting performance, according to Check Point’s 2026 Cloud Security Report. →Voice agents are easy to talk into things. Researchers got a leading realtime voice model to give harmful answers to about 70 percent of spoken requests by varying their speed, pitch and background noise →A transcript check can be blinded. An attacker can speak the instruction and tamper with the recording, so the transcript shows garbled words or nothing at all while the model still receives the original audio →Audio guardrails screen the sound itself. They check […]

The post Audio Prompt Injection: What Voice Agents Hear That Text Guardrails Miss appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/JYHq8dF
via

HTTP2Shell: The Brotli Vulnerability That Extends Beyond Node.js

The key point: CVE-2026-111497 now increasingly referred to as HTTP2Shell and initially reported as Node2Shell, is broader than a Node.js vulnerability.  The vulnerability is associated with Brotli decompression, so defenders need to find vulnerable implementations across applications and infrastructure, confirm whether untrusted compressed data can reach them, and patch the affected component. AT A GLANCE ·       CVE: CVE-2026-111497 ·       Affected layer: Brotli decompression implementations used by multiple runtimes and infrastructure components ·       Primary risk question: Can attacker-controlled Brotli-compressed data reach a vulnerable decoder? ·       Immediate priority: Discover, assess reachability, patch, redeploy, and hunt Context HTTP2Shell emerged shortly after PHP2Shell, but […]

The post HTTP2Shell: The Brotli Vulnerability That Extends Beyond Node.js appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/H3xmRNj
via

Three Gartner Reports. One Emerging Direction for Exposure Management.

In a short period, Check Point has appeared across three new Gartner research publications examining autonomous remediation, exposure assessment, and the future of preemptive cybersecurity. These three reports ask very different questions. But how do they connect? One examines how organizations can trust autonomous remediation. Another looks at the growing exposure management market and the increasing importance of exploitability. The third looks further ahead at the technologies Gartner believes will shape preemptive cybersecurity. Taken together, the reports capture a broader change already underway across the industry. Exposure management is expanding from identifying risk toward understanding exploitability and enabling action. Exploitability […]

The post Three Gartner Reports. One Emerging Direction for Exposure Management. appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/Mt6rb58
via

Yoba Malware Abuses Signed ChatGPT and Claude Binaries to Target AI Users

As interest in new AI tools accelerates, cybercriminals are finding ways to exploit the trust surrounding them. Check Point Research (CPR) has uncovered Yoba, a malware campaign using fake GitHub projects impersonating supposed ChatGPT and Claude desktop applications, while abusing legitimate, signed OpenAI and Anthropic executables to help load malware. The campaign shows how the rush to access new AI tools is creating opportunities for social engineering, even when the AI platforms themselves have not been compromised. Key Findings Attackers used fake GitHub projects impersonating supposed ChatGPT and Claude desktop applications to distribute Yoba malware. Legitimate, signed OpenAI and Anthropic […]

The post Yoba Malware Abuses Signed ChatGPT and Claude Binaries to Target AI Users appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/jonXm5P
via

Rising Borrowing Costs Shouldn’t Delay Your Security: Meet Firewall-as-a-Subscription

Why more companies are moving from hardware purchases to one predictable subscription. Attacks don’t wait for budget approval. But firewall purchases often do, and that wait is getting longer. Rising interest rates have made borrowing more expensive. For a midsize company, a single rate rise can cost the equivalent of a branch office firewall. A large hardware purchase now means a harder conversation with finance and a longer approval cycle. Meanwhile, the old firewall stays in place; the network grows, and so does the risk. Security is urgent. The way organizations pay for it isn’t keeping pace. Check Point Firewall-as-a-Subscription […]

The post Rising Borrowing Costs Shouldn’t Delay Your Security: Meet Firewall-as-a-Subscription appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/YrRQABw
via

Infostealers Are Targeting AI: What Aokigahara Infostealer Reveals About the New Developer Attack Surface

AI assistants have quickly become part of everyday work, particularly for developers. Tools such as ChatGPT, Claude, Cursor and GitHub Copilot increasingly sit alongside source code, cloud environments and other sensitive business resources, making the credentials and sessions connected to these tools increasingly important to protect. New research from Check Point Research (CPR) into the Aokigahara infostealer shows how attackers are adapting to this shift. The malware specifically collects authentication files, session data and configuration files associated with popular AI applications and development tools. The significance goes beyond gaining access to an AI account. If stolen credentials or session tokens […]

The post Infostealers Are Targeting AI: What Aokigahara Infostealer Reveals About the New Developer Attack Surface appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/RwyB8eX
via

PHP2Shell: Log4Shell at AI Speed — A CISO’s Perspective

When the PHP2Shell vulnerabilities became public, my thoughts went back to Log4Shell.  Five years ago, Log4Shell showed how difficult it can be to respond to a critical vulnerability in software used across countless applications and services. PHP2Shell presents a similar challenge. The technical details are different, but the need to quickly understand where you’re exposed and take action is very familiar.  I believe we should expect more incidents like this. AI is making it possible to discover vulnerabilities in widely used software at a speed and scale we haven’t seen before. As these capabilities become more accessible, we are likely […]

The post PHP2Shell: Log4Shell at AI Speed — A CISO’s Perspective appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/RD7HKeV
via

PHP2Shell Update: One in ten organizations impacted as Europe bears the brunt

Two critical vulnerabilities in PHP, the programming language behind a large share of the world’s websites, online stores and internet services, are being actively targeted. Since details of CVE-2026-109618 and CVE-2026-109619, known as PHP2Shell, leaked ahead of planned disclosure, 10% of organisations worldwide have been impacted (“impacted” means exploitation attempts, we have no way of knowing if the attack has been successful), according to Check Point Research. Hourly activity peaks at midnight, when fewer security teams are watching. Government institutions in the Netherlands and Finland have taken some of their websites offline as a precaution. What PHP2Shell is and why […]

The post PHP2Shell Update: One in ten organizations impacted as Europe bears the brunt appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/Cp4xzSW
via

Custom Guardrails: The AI Rules No Vendor Can Write for You

Key takeaways Only 14 percent of organizations say they have formal GenAI policies that are actively enforced and audited, according to Check Point’s 2026 Cloud Security Report. →The costliest AI mistakes break rules no vendor ships. Air Canada was held liable for a refund its chatbot promised and the airline never offered →The usual fixes fall short. Models can be argued out of system prompts, keyword lists miss meaning, and US-centric DLP lists miss the data your regulators protect →Custom guardrails turn a sentence into an enforced rule. Describe it, test it in under five minutes, and run it on […]

The post Custom Guardrails: The AI Rules No Vendor Can Write for You appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/MroZgi0
via

How Fraudulent Airline Support Numbers Exploit Search Results and AI-Generated Answers

A canceled flight. A missed connection. A refund that has yet to arrive. When travelers need urgent help, they often turn to search engines, social media, or AI tools to find a customer support number. Threat actors are increasingly exploiting that moment of urgency. In our previous investigation, Scammers Are Watching Airline Complaints and Posing as Customer Support, Check Point Exposure Management researchers uncovered scammers monitoring airline complaints on social media and impersonating customer support representatives. Check Point Exposure Management researchers identified a related campaign that targets an earlier point in the support journey: the search for help itself. Threat […]

The post How Fraudulent Airline Support Numbers Exploit Search Results and AI-Generated Answers appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/QSKkhqR
via

As AI Is Reshaping the Data Center, Security Has to Keep Up.

Modernizing the Data Center for the AI Era For years, the data center was treated as one of the most controlled parts of the enterprise. Critical applications and sensitive data lived behind a hardened perimeter, where access could be closely managed and where traffic followed relatively predictable paths. That environment has changed. Quickly. AI training and inference workloads, private models, GPU infrastructure, containers, APIs, and autonomous agents are joining the core applications and services already operating inside the data center. According to the 2026 Check Point Cloud Security Report, 29% of organizations are already moving AI workloads to on-premises or […]

The post As AI Is Reshaping the Data Center, Security Has to Keep Up. appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/ojh3rI6
via

Keeping AI Agents on Task: Check Point Advances AI Agent Safety with NVIDIA OpenShell Reference Architecture

How Check Point and NVIDIA collaborate an agent governance, control over how they run, and in-silicon security built into the AI infrastructure  Agents can go wrong without an attacker  AI agents now plan, use tools, work with sensitive data and hand work to other agents. Their value comes from doing things nobody scripted step by step. That also creates a new kind of risk: an agent can follow every rule and still do the wrong thing.  No attacker is needed. An agent can:  Access sensitive data, credentials, or internal systems  Absorb untrusted or misleading content as if it were a […]

The post Keeping AI Agents on Task: Check Point Advances AI Agent Safety with NVIDIA OpenShell Reference Architecture appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/UVD36AT
via

The Risk of AI-Powered Antidetect Browsers in Online Fraud and Influence Operations

Online platforms rely on digital signals to identify suspicious accounts and activity. Antidetect browsers make those signals harder to trust by allowing operators to create and manage large numbers of browser profiles that appear to belong to different users, devices and locations. While these tools have legitimate uses, Check Point Research found documented evidence of their use in criminal ecosystems for activities including account takeover and payment fraud. AI and automation can further increase the scale of these operations. Key Findings Antidetect browsers allow operators to manage large numbers of isolated browser profiles that appear to come from different users, devices […]

The post The Risk of AI-Powered Antidetect Browsers in Online Fraud and Influence Operations appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/83npe56
via

Check Point Expands Strategic Collaboration with AWS to Accelerate AI and Cloud Security Capabilities

Check Point and Amazon Web Services (AWS) have worked together to help organizations innovate in the cloud while enhancing security. As a global cyber security leader, Check Point gives AWS customers a practical way to strengthen protection across cloud networks, applications, workloads, and emerging AI environments. Available through AWS Marketplace, Check Point solutions are designed to help customers move faster on AWS while extending consistent, prevention-first security across the services they rely on every day. In 2026, that collaboration is expanding again, with a focus on helping customers secure AI initiatives built on AWS. As enterprises move generative AI from […]

The post Check Point Expands Strategic Collaboration with AWS to Accelerate AI and Cloud Security Capabilities appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/BWEIraM
via

Inside the Q3 2026 Brand Phishing Report: Microsoft Now Accounts for One in Three Brand Impersonation

In Q3 2026, Microsoft was the most impersonated brand in phishing attacks, accounting for 33% of all brand impersonation attempts tracked by Check Point Research, up from 23% in Q2. Attackers favor Microsoft because stolen Microsoft credentials can unlock an entire corporate environment, including email, files and collaboration tools, while many of the other top brands mainly lead to personal accounts. Apple, Google, LinkedIn and Amazon completed the top five, and together these brands made up more than half of all attempts, showing how heavily attackers lean on a small group of trusted names. This quarter also brought examples showing […]

The post Inside the Q3 2026 Brand Phishing Report: Microsoft Now Accounts for One in Three Brand Impersonation appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/5wEXe9I
via

Securing AI Agents from Discovery to Runtime: Check Point on Google Cloud’s Gemini Enterprise

AI agents are moving into production fast. They read enterprise data, call tools, and take actions across business systems. That creates a new problem: an agent can be fully authenticated and authorized and still be manipulated into doing the wrong thing. Google Cloud’s Gemini Enterprise gives enterprises a central control plane for agents. Check Point extends that control plane with AI-specific risk assessment and real-time, context-aware protection. This post details where Check Point integrates and what it looks like for a security team. Google Cloud’s direction with agent orchestration, registries, and gateway layers signals a clear design pattern: centralized AI […]

The post Securing AI Agents from Discovery to Runtime: Check Point on Google Cloud’s Gemini Enterprise appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/Z7cqTA0
via

Fake Calendar Invites Target Thousands of Organizations in Microsoft Credential-Phishing Campaign

A calendar invite is one of the most routine messages in the modern workplace. It lands in the inbox, asks for a few seconds of attention, and is rarely questioned. Attackers know this, and they are now using fake meeting invitations to steal Microsoft account credentials. We have identified several large-scale phishing campaigns that abuse fake calendar invites to harvest credentials. Some arrive as an .ics attachment (example 1). The file is valid calendar data, so the attack abuses calendar features and user trust without exploiting a software vulnerability. Others use a link that appears to lead to a different […]

The post Fake Calendar Invites Target Thousands of Organizations in Microsoft Credential-Phishing Campaign appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/01ySJPo
via

Researchers Uncover Thousands of Phishing Domains Hiding Behind Inactive Pages

A domain returns a blank page. Maybe a generic “Not Found” message. There is no login form, suspicious branding, or obvious phishing content to investigate. Yet the phishing site may still be there. Check Point Exposure Management researchers uncovered a large scale, multi region phishing infrastructure designed to conceal active campaigns behind seemingly inactive domains. Starting with a handful of phishing URLs, the investigation identified a repeatable technical fingerprint that ultimately connected the activity to thousands of rotating domains targeting organizations and users across multiple regions. The infrastructure uses regional paths, localization, session controls, and token gated access to determine […]

The post Researchers Uncover Thousands of Phishing Domains Hiding Behind Inactive Pages appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/2BJVzWF
via

Dangerous Deviation Detection: Catching the Agent Action Nobody Asked For

Key takeaways In the past year, 78 percent of organizations reported a confirmed or suspected AI-related security incident, according to Check Point’s 2026 Cloud Security Report. →No attacker required. Agents plan their own steps, and an agent trying to help can take one nobody asked for that deletes data, leaks it or widens someone’s access →Permissions and prompt screening miss the action itself. Least privilege cannot take away access the job needs, and prompt attack detection screens what goes into the agent, not what it is about to do →Dangerous Deviation Detection checks each tool call before it runs. It […]

The post Dangerous Deviation Detection: Catching the Agent Action Nobody Asked For appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/5fxm81I
via

Election Scams Are Getting Harder to Spot as Real Campaign Outreach and Fraud Converge

In September, viewers of a Pennsylvania TV station received a text telling them to “verify your voter profile.” The station’s investigator could not find out who sent it, and the link was flagged as malware. When Check Point Exposure Management researchers followed the link, it led to a political fundraising page. That small episode sums up the main finding of our report on 2026 US midterm election threats. Voters are having a hard time telling which messages are real, and attackers benefit from that doubt. Real outreach and fake outreach look alike Between July and September, election offices in Rhode […]

The post Election Scams Are Getting Harder to Spot as Real Campaign Outreach and Fraud Converge appeared first on Check Point Blog.

from Check Point Blog https://ift.tt/Mu1D3re
via