Microsoft says Teams will begin using AI-powered phishing detection in early November 2026 to detect malicious behavior in messages. The rollout is part of a targeted release phase expected to end by mid-November, with the new security feature likely reaching general availability worldwide between early and mid-December. Microsoft first revealed in a Microsoft 365 Roadmap entry last week that the AI-powered phishing detection will use conversation context (without retaining message content) to identify phishing and suspicious behavior in higher-risk messaging scenarios, such as external first-contact chats. "Microsoft Teams is introducing AI-powered, context-aware phishing detection to help identify phishing, social engineering, impersonation attempts, and deceptive intent," the company said in a Microsoft 365 Message Center update on Wednesday.
PHP warns of a new, unlikely code execution flaw, CVE-2026-115560, and advises disabling session upload progress until a fix is released. PHP has warned of a new code execution vulnerability, tracked as CVE-2026-115560, just two days after releasing emergency patches for the critical PHP2Shell flaws. The vulnerability affects only a specific PHP configuration that PHP says is unlikely to be used in practice. An application must meet several conditions: it must configure session.serialize_handler as php_binary or php_serialize, change it to php at runtime using ini_set(), and make this change before calling session_start(). Session upload progress must also remain enabled. In an advisory published on October 23, the project recommended disabling session upload progress until a fix becomes available in a regular release. The vulnerability is separate from PHP2Shell and is not fixed by the October 21 updates for PHP versions 8.2.35, 8.3.36, 8.4.27 and 8.5.12. “The issue can only occur under an unusual combination of configuration and application code. An application must set session.serialize_handler to php_binary or php_serialize in its configuration and then switch it to php at runtime with ini_set() before calling session_start(). Session upload progress must also still be enabled.” reads the advisory. “Applications that never change the serialize handler at runtime can keep the feature enabled. Since this setup is unlikely in practice, we are handling it in the regular release cycle and will publish a full advisory for CVE-2026-115560 with the next release.” Applications that do not change the serialization handler while running are not exposed through this attack path, even if session upload progress is enabled. PHP recommends a simple temporary fix: set session.upload_progress.enabled=0 until the next routine bugfix release. This is the same mitigation already recommended in the October 21 advisory for CVE-2026-109619. Administrators who applied this setting and have kept it in place do not need to take further action for this issue. Session upload progress is enabled by default and allows applications to track the progress of file uploads. Disabling it reduces exposure to this specific vulnerability path. PHP will fix CVE-2026-115560 through its regular release cycle because the vulnerable setup is uncommon. It plans to publish a full advisory with the next release. PHP has not yet identified affected versions, confirmed when the fix will arrive, assigned a severity score, or explained whether the flaw shares a root cause with CVE-2026-109619. No attacks targeting this new issue have been reported. On October 21, PHP released emergency patches for two critical vulnerabilities tracked as CVE-2026-109618 and CVE-2026-109619 (CVSS score of 9.2), affecting every PHP version that handles web requests. The bug code dates back to PHP 5.4 from 2012, lives in the PHP engine itself rather than in any application, and allows unauthenticated remote code execution via crafted HTTP requests. Cloudflare researchers observed tens of thousands of exploit attempts within 45 minutes of public disclosure, hundreds of thousands within two hours, and millions by October 22. Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, PHP)
Microsoft says Windows Server 2016 will stop receiving security updates after it reaches the end of support in January 2027. Windows Server 2016 was released in September 2016 at Microsoft's annual Ignite conference and reached general availability one month later, on October 12, 2016 It reached the end of mainstream support in January 2022, and extended support will end in roughly three months, on January 12. "The January 2027 Windows security update will be the last update available for this version. After this date, devices running Windows Server 2016 will no longer receive monthly security updates containing protections from the latest security threats," Microsoft said in a message center update. "This end of updates in 2027 applies to all editions of Windows Server 2016: Datacenter, Essentials, MultiPoint Premium and Standard."
The PHP security team has shared mitigation measures for a new remote code execution vulnerability that affects systems with non-default configurations and has no patch. PHP is an open-source web development language used by nearly 70% of all websites on the Internet. According to VirusTotal estimates, more than 5.1 million PHP hosts are exposed online, and roughly 3.8 million have at least one port open to attacks. According to a Friday advisory, this security flaw (tracked as CVE-2026-115560) can only be exploited when targeting systems "with an unusual combination of configuration and application code." "An application must set session.serialize_handler to php_binary or php_serialize in its configuration and then switch it to php at runtime with ini_set() before calling session_start(). Session upload progress must also still be enabled," the PHP team explained.
AI agents can be manipulated through malicious emails, websites, and plugins into executing commands, stealing data, or deleting files using their legitimate permissions. Gen Digital explains how its Agent Trust Hub helps protect agents through skill verification, secure connectivity, and runtime controls that intercept dangerous act
Hewlett Packard Enterprise (HPE) released security updates to patch 72 vulnerabilities in the Aruba Operating System (ArubaOS), including eight maximum-severity authentication bypass and code execution issues. Also known as AOS, ArubaOS powers Aruba Networking enterprise wireless infrastructure, including mobility conductors, access points, controllers, and gateway devices. The eight maximum-severity security flaws patched today affect multiple AOS-8 and AOS-10 software versions, and they can be exploited by remote threat actors without privileges through low-complexity attacks that don't require user interaction. HPE warned customers to patch these security issues as soon as possible and provided mitigation measures, including restricting access to AOS management interfaces. "Customers are strongly urged to patch their instances due to the complexity, breadth, and impact of these vulnerabilities," the company warned in a Thursday security advisory.
A New Mexico judge ordered Meta to pay $25 billion for misleading users about privacy, setting a record penalty. Meta plans to appeal. This week a New Mexico judge ordered Meta to pay the state $25 billion. The order follows a jury finding that the company deceived Facebook users about their privacy, and it’s the biggest fine ever imposed on Meta. State District Court Judge Francis J. Mathew wrote that Facebook’s deception was intentional and that executives consistently misled users about how the platform worked. “Facebook promised users that they (the users) controlled their data. This was false,” he said, as reported by the Associated Press. This is a case about promises, not about a breach. Jurors found in September that Meta committed over 43 million violations of New Mexico consumer protection laws by making misleading public statements about how it handled users’ personal data. The case grew out of the Cambridge Analytica scandal, in which a political consulting firm harvested Facebook data to build targeted political ads for the 2016 election. New Mexico had asked the court to fine Meta between $35 billion and $40 billion, arguing that only a large penalty would push the company to change its behavior. Meta’s lawyers said the fine should not exceed $3.4 billion and argued that any higher amount would be unconstitutional. The jury found more than 43 million violations, including 2.1 million under specific claims. Each violation can carry a fine of up to $5,000. New Mexico Attorney General Raúl Torrez said the ruling sends a clear warning to technology companies that they must follow the law. Under the state’s constitution, fines for consumer protection violations go toward education. However, the money may remain tied up while Meta appeals the decision. Meta plans to appeal the $25 billion ruling. The company argues that the decision and the way New Mexico is enforcing consumer protection laws make it almost impossible to operate in the state. Meta’s chief legal officer, C.J. Mahoney, says the case is an attempt to restrict free speech and impose new rules on what people can say online. The company says it is confident it will win the appeal. Meta’s lawyers also argued that the state failed to prove that residents had actually seen the misleading statements or suffered harm as a result. Meta is worth $1.8 trillion and brought in more than $200 billion in revenue last year, and its shares rose 2% after the announcement, with earnings due that day. A nonprofit director called $25 billion a drop in the bucket for Meta, though he said it shows people want accountability. Both can be true: the company says doing business in New Mexico is all but impossible, and Wall Street shrugged. The history explains why the number is so large. Cambridge Analytica got data through a personality quiz app that around 300,000 people took, and the app also pulled data on their friends, building a dataset of 87 million profiles. Meta banned the app in 2015 and asked Cambridge Analytica to delete the data, and the firm shut down in May 2018. The company later agreed to pay $725 million in December 2022 to settle a class action over sharing with Cambridge Analytica, and the FTC fined it $5 billion in 2019, a fifth of today’s award. New Mexico was not part of the $18 billion settlement Meta reached in August with almost all other US states and territories. The deal includes stronger child-safety measures and $459 million related to the Cambridge Analytica scandal. This is Meta’s second legal defeat in New Mexico in six months. In March, a jury ordered the company to pay $375 million for misleading users about children’s safety. Meta’s appeal could reduce the $25 billion fine if a court lowers the number of violations. The case also sends a warning to companies: privacy promises have legal consequences. As the judge stated, Facebook told users they controlled their data, but this was not true. Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, Cambridge Analytica Scandal)
China-linked Storm-0940 used AI tool ARTEX to target 2,250 Microsoft 365 tenants, exploiting stolen ConnectWise credentials at machine speed. Microsoft published a detailed technical report naming Storm-0940, a China state-aligned group, as the actor behind the ConnectWise Cloud Backup supply chain compromise. The group used ARTEX, an open-source AI-driven penetration testing framework, to run attacks at machine speed against hundreds of organizations across aerospace, defense, energy, financial services, government, manufacturing, and more. “Several of these Storm-0940 ARTEX-driven attacks have resulted in the actor obtaining privileged access to an organization’s cloud infrastructure – including multiple instances where the time from initial targeting to privileged access was less than 3 hours.” reads the report. “Microsoft assesses that Storm-0940 was responsible for the recently disclosed ConnectWise supply chain compromise of a multi-tenant application.” Storm-0940 has been active since 2021 and has historically relied on password spraying and OAuth application abuse to get into Microsoft 365 environments. Adding ARTEX changed the scale of what they could do. In some attacks, the time between initial targeting and achieving privileged cloud access was under three hours. Before obtaining the ConnectWise credentials, Storm-0940 launched a password spraying attack between October 10 and 14. The attack came from the IP address 144.172.64[.]99. Microsoft recorded more than 151,000 failed login attempts against 327 accounts at eight organizations. Activity peaked on October 11, with 2,900 failed attempts per minute. Forty-eight accounts each received more than 1,000 failed attempts. None of these attempts succeeded. The situation changed on October 15, one day after the ConnectWise breach became public. The AI agents stopped trying passwords and switched to a different attack method. By then, Storm-0940 had obtained the credentials of several ConnectWise cloud application accounts. The agents used these credentials to log directly into customer Microsoft 365 environments. “Storm-0940’s use of ARTEX accelerated the speed of compromised application abuse to authenticate to Microsoft 365 customer environments through applications that had already been granted access.” Microsoft continues. “This allowed the threat actor to operate without requiring an interactive user sign-in. “ The agents used applications that already had permission to access their targets. This allowed the attackers to enter customer environments without requiring users to sign in interactively. In a subset of environments where authentication succeeded, the agents didn’t stop there. They moved into Microsoft Graph and ran directory reconnaissance, modified application credentials to establish persistence, and attempted changes to identities, groups, and privileged roles. Microsoft identified multiple customer environments with evidence of this follow-on activity. The persistence mechanism is the part defenders need to act on now, even if they’ve already disabled the ConnectWise connections. “These changes can remain active as a backdoor even after the original compromised application is mediated.” says the IT giant. An attacker who added their own credentials to a trusted application before you revoked access still has a working key. Reviewing all service principal credential additions since at least October 1 is the right starting point, not October 15 when the public breach was announced. ARTEX is not a tool used only by Storm-0940. Microsoft says threat actors in several regions have used the same framework. Originally open-source, ARTEX was written in Go and can use different AI models to plan and carry out tasks. Researchers at Zscaler and Recorded Future also found a Chinese-speaking group using it to target South Korean financial institutions. The setup used DeepSeek, GLM, and Grok as coordinated AI agents, with GLM checking DeepSeek’s work. After these attacks became public, the project went closed-source. However, anyone who already has a copy can still use it. The speed point is what Microsoft keeps coming back to in its report. Agents don’t take weekends, don’t sleep, and can run multiple attack paths simultaneously. “This activity demonstrates how threat actors are integrating agentic AI into existing offensive security workflows and allows them to scale their operations exponentially.” Microsoft concludes. “Agents are not constrained by factors affecting humans – they can work around the clock, on weekends, and on holidays and attempt multiple breach paths simultaneously enabling significantly reduced time from selecting a target to achieving objectives, increasing the speed and impact of complex attacks like supply chain compromises.” When Storm-0940 had valid application credentials, they didn’t have to iterate slowly. They hit 2,250 tenants with 2,900 sign-in attempts per minute, a pace at which manual detection and response simply can’t keep up. Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, ARTEX)
Hackers compromised the LiteLLM project’s PyPI package and published malicious Docker images with a payload that steals credentials, secrets, and environment variables. The supply-chain attack affected all PyPI wheels (.whl installation packages) for LiteLLM version 1.107.0, impacting Windows, Linux, and macOS systems. LiteLLM is an open-source Python SDK and AI gateway that enables applications to access AI models from different providers through a common API. It is a massively popular project with 61,800 GitHub stars, 92 million monthly downloads on PyPI, and is used by high-profile companies like NVIDIA, Netflix, AT&T, and Okta. The project says that the release pipeline was compromised on October 21 and a threat actor pushed version 1.107.0 to the Docker registry and PyPI. "This version contained compiled malware which steals credentials and loads remote-code execution tools," the maintainer says, adding that customers running the private enterprise were not impacted.
Attackers are exploiting a zero-day SQL injection flaw in Arista's VeloCloud Orchestrator (VCO), the central management system for VeloCloud SD-WAN networks, the company said in an advisory on October 22. The flaw lets anyone who can log in to the orchestrator, even with a read-only customer account, gain operator privileges, the top level of control over it. For
Confirm this action.
Leaving now will discard your changes.