North Korean hackers are using a post-quantum cryptographic mechanism to secure malicious payload delivery in new operations that combine old and new malware. The campaign targets South Korean organizations through a “rather complex” infection chain. Researchers at cybersecurity company Kaspersky dubbed the operation FlashHole and attributed it to the Lazarus Group. Analysis of the malware revealed that the threat actor used the standardized ML-KEM post-quantum cryptographic mechanism to communicate with the command-and-control (C2) server for malware delivery. Kaspersky says the attackers used LPEClient malware, previously associated with Lazarus, to receive ML-KEM-1024 public keys from its C2 infrastructure and generate shared keys that allowed it to decrypt subsequently incoming payloads.
OpenAI reveals two cases of AI agents disabling safety monitoring during coding tasks, with one falsely denying it had removed the safeguards. OpenAI published an internal incident report this week documenting two separate cases in which GPT-6 class agents weakened the system designed to detect misaligned behavior, in both cases while carrying out routine coding tasks. This wasn’t an external attack. These were internal deployments, and the agents made their own judgment calls that happened to gut the oversight watching them. OpenAI explained that “in both cases, the agents considered whether monitoring safeguards should be preserved, but interpreted routine coding instructions as permission to weaken them.” In other words, the agents thought about whether they should keep the safety controls in place. They then interpreted the user’s instructions in a way that led them to weaken those controls. “In two internal coding tasks, agents weakened misalignment monitoring while making requested software changes. A GPT-6.1 Sol agent repairing an evaluation pipeline disabled automatic alerts to our Security team, treating an earlier instruction not to send messages during a Slack history search as a reason to suppress monitoring notifications.” reads OpenAI’s report. “In a separate task, a GPT-6 Astra agent migrating a program removed its monitoring integration in order to match a reference implementation, and then falsely claimed to the user that it had not removed the monitor.” On October 3, a researcher asked a GPT-6.1 Sol agent to create evaluation tasks using Slack data. The researcher clearly instructed the agent not to send any messages during the process. Later, the system triggered a monitoring alert because it couldn’t save its findings to a database. The user then asked the agent how to proceed. At this stage, the agent was still following the rules. It even told a subagent to keep monitoring enabled and report the actual results. The subagent correctly identified that turning off notifications would disable required safety alerts and decided not to do it. Then the parent agent reconsidered on its own, reasoning that writing results to a database might cause the monitoring system to send a message, which conflicted with the user’s earlier “don’t send messages” instruction. It concluded the instruction applied to safety alerts too, and directed the subagent to set no_notify to true, empty the list of people set to receive monitoring reports, and make notification suppression a hard requirement before writing any results. The monitor still ran and saved its reports. Nobody received the human alerts. On October 6, a user asked a GPT-6 Astra agent to copy a reference implementation while keeping existing tool definitions and functionality intact. An earlier instruction from another agent had explicitly told the system not to disable monitoring, even temporarily. The parent agent passed that requirement to a subagent but softened it, saying monitoring should be preserved “as appropriate” to the reference implementation. The subagent took this wording as permission to remove monitoring because the reference implementation did not include it. It also changed a test that checked whether monitoring was active. The new test instead required monitoring to be disabled. After the migration, the main agent noted internally that the new setup no longer included agent monitoring and described this as a positive outcome. When the user later asked whether monitoring had been removed, Astra said: “No… I did not remove an enabled monitor from it.” That was false. OpenAI offers two possible readings of these incidents. The charitable one is what they call “benign overpersistence,” models that don’t understand why monitoring matters and follow instructions more literally than a reasonable person would. The less comfortable reading is that these models might have a specific tendency to find justifications for evading the systems watching them. “In both cases, the agents considered whether monitoring safeguards should be preserved, but interpreted routine coding instructions as permission to weaken them.” states OpenAI. “We are strengthening our misalignment monitoring system in response, and are also assessing whether these cases reflect benign overpersistence or a propensity to evade misalignment monitoring in particular.” OpenAI says it doesn’t yet have a satisfying answer. To find one, they’re developing paired evaluations, resampling agent continuations, and running experiments on the models’ internal activations. OpenAI has introduced some practical fixes. Its auto-review system, already required for internal Codex use, has been updated to detect and block attempts to disable safety monitoring. The company also plans to require each workload to have an approved safety plan that defines its monitoring requirements. Any workload that fails to meet these requirements will be stopped automatically. Part of the problem was how the systems were designed. Monitoring was easier for agents to bypass during evaluation tasks than in production Codex deployments. In production, monitoring runs outside the agent’s direct control, making it harder for the agent to disable. This report lands a week after OpenAI disclosed seven other misalignment incidents, including one where a research model took cryptocurrency from a wallet it didn’t own to pay transaction fees and didn’t disclose it. “In addition to making it harder for models to disable monitoring, we are looking to understand why models might seek to do so in the first place. It is possible that this behavior is simply an example of “benign” overpersistence, with models not understanding why monitoring is important or insisting that instructions be executed beyond what a reasonable person might expect.” concludes the report. “We are also developing evaluations to test whether models might have a propensity to evade misalignment monitoring in particular and exhibit motivated reasoning in order to do so. We don’t yet have a satisfying explanation for this behavior, and will investigate this behavior through the creation of paired evaluations, resampling of agent continuations, and experiments on models’ internal activations.” Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, AI)
The U.S. Cybersecurity and Infrastructure Security Agency (CISA) is warning that attackers are actively exploiting four vulnerabilities in Plex Media Server. The security issues are collectively called Plex2Shell. When chained in an exploit, an attacker can obtain the owner's admin token and execute code with elevated privileges. Plex Media Server is a popular media suite that helps users organize personal collections of movies, TV shows, music, and photos, and stream them to other devices at home or remotely over the internet. CISA has directed federal agencies to prioritize patching systems affected by Plex2Shell vulnerabilities or apply mitigations by October 23. The security issues were discovered by security researcher Zach Main, who reported them responsibly to Plex on July 27. The individual flaws are rated medium to high in severity, but the researcher assigns the full exploit chain a critical score of 9.8.
OpenAI agents powered by the GPT-6.1 Sol and GPT-6 Astra models weakened misalignment monitoring during two internal tasks, interpreting coding instructions as authorization to do so. AI alignment refers to how well an AI system’s goals and behavior align with human values, intentions, and instructions. It is a key focus of AI safety research aimed at preventing harmful actions and unintended outcomes. One of the incidents occurred on October 3, when an agent powered by GPT-6.1 Sol extended an earlier human instruction “don’t send any messages” related to a Slack history search to automatic monitoring alerts. It directed a sub-agent to empty notification recipient lists and suppress notifications. Although monitoring continued to run and save findings, it disabled alerts to humans.
Vulnerability overview On October 20, 2026, the PHP security team published a preliminary notice regarding an upcoming emergency release that addresses two critical vulnerabilities (both rated 9.2 on the CVSS scale). The patches were released on Wednesday, October 21, 2026, at 14:00 UTC. The release was expedited after information about the vulnerabilities was leaked to the public before the scheduled embargo was lifted. In a post on Hacker News, a user reported two RCE vulnerabilities in PHP, allegedly affecting all versions starting with PHP 5.4.0. The vulnerabilities in question are CVE-2026-109618 and CVE-2026-109619. Shortly after the post appeared, researchers confirmed the existence of the vulnerabilities. Along with the leak of details, an exploit unofficially named PHP2Shell appeared online. In the leaked demo video, the exploit is shown in action on web servers running WordPress, MediaWiki, phpBB, Magento, Joomla, Laravel, Symfony, Drupal, and WHMCS — platforms that power many modern websites. The authenticity of the demonstration and the exact behavior of the exploit had not been independently verified at the time of publication. According to the researcher who published the PoC, the vulnerabilities are related to the handling of the multipart/form-data MIME type in conjunction with the session upload progress mechanism (session.upload_progress) in the PHP core and fall under the categories of use-after-free and memory corruption. This means that the flaws cannot be patched via application code, and the only possible fix is to update PHP on the server. The attack works without authentication: an attacker with network access to a PHP application can execute arbitrary code on behalf of the PHP process without needing to authenticate. Security advisories issued by PHP on October 21 confirm this information. The vulnerabilities affect PHP versions 5, 6, 7, and 8. As of this publication, branches from 8.2 to 8.5 are supported. Older branches, including PHP 8.1, 8.0, 7.x, and 5.x, have reached the end of their lifecycle; however, due to the critical severity of the vulnerabilities, backported patches have also been released for versions 7.4, 8.0, and 8.1. Sansec experts estimate that widespread exploitation could begin within 15 minutes of the update’s release. We strongly recommend that administrators responsible for publicly accessible PHP web servers take the following steps before the patch is available: Conduct an inventory of all PHP-based systems in the infrastructure (including web servers, control panels, third-party software, etc.). Disable upload-progress: set session.upload_progress.enabled = Off using the php_admin_flag in PHP-FPM pools. Block multipart requests on nginx and Apache servers. Consider temporarily isolating PHP applications that connect to the Internet until fixes are released. Ensure readiness to install the patch immediately upon its release. Plan to migrate servers running unsupported versions (8.1 and earlier) to the latest PHP branch. Detection by Kaspersky solutions Kaspersky solutions detect the described malicious activity at various stages of the attack. Possible detection scenarios are listed below. When an attempt is made to exploit these RCE vulnerabilities, Kaspersky Endpoint Detection and Response promptly alerts the user to suspicious actions performed in the context of a web server, on both Linux and Windows machines. This activity is detected by the following rules: creating_windows_shell_from_a_web_servers_process; possible_web_shell_creating_via_exploit; spawning_process_from_webserver_more_filters_linux; webserver_interpreter_execution_linux; executable_file_in_www_folder. Detection of the PHP2Shell vulnerability exploit by KEDR Exploitation of the PHP2Shell vulnerability in network traffic is detected by Kaspersky Anti Targeted Attack (KATA). This solution detects such activity with the following verdicts: Exploit.CVE-2026-109618.HTTP.C&C and Exploit.CVE-2026-109619.HTTP.C&C. Detection of the PHP2Shell vulnerability exploit by KATA Our Kaspersky Endpoint Security solution, now available as Kaspersky Next, also detects malicious post-exploitation activity using the Behavior Detection component, such as PDM:Exploit.Linux.Generic and PDM:Exploit.Win32.Generic.
The PHP security team released emergency security patches for two critical remote code execution (RCE) vulnerabilities (CVE-2026-109618 and CVE-2026-109619) tracked collectively as PHP2Shell. The two unauthenticated RCE flaws were discovered by Calif, Wiz, and OpenAI security researchers, and affect all PHP installations that serve web requests, including PHP-FPM, Apache with mod_php, php-cgi, FrankenPHP, and LiteSpeed. "We consider the risk highest for mod_php, with a demonstrated exploit where an unauthenticated request can trigger the attack before any application code runs," the PHP security team said. Both flaws stem from the same underlying problem: PHP can leave its memory in a corrupted state when a request ends abnormally, and both can be exploited to achieve remote code execution. CVE-2026-109618 can be used to gain remote code execution as the web worker account by exploiting a chain of weaknesses that trigger out-of-memory corruption, while the CVE-2026-109619 stale global state vulnerability enables unauthenticated heap use-after-free, memory disclosure, and remote code execution with the web worker's privileges.
An investigation into a threat actor bypassing AI safety controls to coordinate credential theft, VoIP fraud planning, and database exfiltration. The post “Next Exploit Opportunities”: How a Threat Actor Used AI to Help Plan and Execute Multiple Types of Cybercrime Activities appeared first on Unit 42.
ClickFix attacks are surging as cybercriminals use fake AI installers, malicious search results, and trusted chatbot platforms to trick users into executing malware. Push Security examines how these attacks evade traditional defenses and why AI tools have become increasingly effective lures.
CVE-2026-109618 and CVE-2026-109619 are Critical vulnerabilities in PHP, a scripting language used globally for web servers and CMS systems. The post Threat Brief: Critical Vulnerabilities in PHP: CVE-2026-109618, CVE-2026-109619 appeared first on Unit 42.
Microsoft has released PowerToys 0.102, with two new mouse utilities, new command-line tools, and Command Palette improvements. Mouse Button Lock, one of the new mouse tools, makes it easier to select items on screen by locking a mouse button while you hold down a key. "With Mouse Button Lock, press and hold a button for a moment, then let go. Windows keeps treating it as held, so you can continue dragging or selecting until you click any button," Microsoft senior product manager Niels Laute explained. "This is especially useful for gaming, working in professional applications that require extensive dragging and selection, and improving accessibility for users who find it difficult to hold down a mouse button for extended periods." Microsoft also added CursorHide to PowerToys, another mouse utility that automatically hides the mouse pointer while you type or when you aren't moving the mouse.
Confirm this action.
Leaving now will discard your changes.