Microsoft announced that IT administrators can now use quick machine recovery on Windows Server 2025 after installing the October 2026 security update. Quick machine recovery (QMR) lets IT admins fix critical boot failures with targeted Windows Update fixes without physical access to affected systems. When enabled, it boots into the Windows Recovery Environment (WinRE) and launches the QMR Tool if Windows 11 fails to start because of a configuration change or a new driver. It then sends crash data to Microsoft, which can remotely fix issues by adjusting settings or removing problematic drivers or updates. "With the October 2026 security update, quick machine recovery extends to Windows Server 2025," the company said in a message center update. "Quick machine recovery is designed to help restore devices experiencing widespread boot issues by using the Windows Recovery Environment (WinRE) to search Windows Update for applicable remediations."
WhatsApp's help pages carried hidden instructions for its support chatbot that told it not to recommend a data protection authority or regulator to users who could no longer delete their accounts, even though WhatsApp's own privacy policy for Europe tells users they can complain to one. The text sat in the pages' code, never shown on screen, until it was removed by October 18. Whether the
A sales manager builds an AI-assisted workflow to route leads. It connects to customer records, summarizes inbound requests, and posts updates to a shared channel. The workflow saves time, reduces manual work, and soon becomes part of the team’s daily routine. Across the same organization, employees are building useful apps, agents, and automations faster than existing processes can support them
Go 1.27.3 and Go 1.26.10, released on October 20, make Go's certificate verifier reject any certificate chain that passes through one of eight intermediate certificate authority keys. Programs built on the new release will refuse connections to sites and services whose certificates were issued under those keys. The Go team says the keys belong to one certificate authority it has not
French cloud and hosting provider OVHcloud is warning users that attackers compromised its open-source command-line interface (CLI) tool to distribute information-stealing malware. The attacker hijacked a developer’s GitHub account to gain access to the OVHcloud CLI repository and modify versions 0.11.0 through 0.15.0 of the tool. OVHcloud offers the CLI tool as an alternative for system administrators and developers to manage cloud and hosting services directly from a terminal, including for automating tasks. OVHcloud CEO Octave Klaba disclosed that the company’s CLI repository on GitHub had been compromised, urging users who downloaded the latest version yesterday to update or remove the tool. Klaba said that the compromised release appeared at 14:47 on October 19 and was detected and blocked at 16:21 after a customer reported it.
Ahold Delhaize, one of the world's largest food retail chains, confirmed that last week's cyberattack disrupted services at stores across Europe and the United States. Right before the cyberattack was disclosed, customers couldn't use websites, apps, and customer service platforms at Albert Heijn, Etos, and Gall & Gall stores in the Netherlands, and Delhaize in Belgium, while employees at the Stop & Shop American regional supermarket chain reported having trouble accessing their schedules and pay stubs. When BleepingComputer asked right after the incident whether it also affected both corporate and store systems and whether any devices were encrypted during the attack, an Ahold Delhaize spokesperson redirected us to the company's statement. In a Monday update, Ahold Delhaize said the cybersecurity incident announced last week forced it to take some systems offline, including its brands' websites and apps, causing service issues, although stores remained open to serve customers.
The Russia-linked state-sponsored threat actor tracked as APT28 (aka BlueDelta, Fancy Bear, and Forest Blizzard) has been attributed to attacks targeting two member states of the North Atlantic Treaty Organization (NATO). The Symantec Threat Hunter Team, which is tracking the activity cluster under the moniker Swallowtail, involved the use of a backdoor called HOOKEDGE and a previously
ConnectWise on Monday said it has taken remediation measures, including strengthening administrative-access protections and resetting accounts and replacing application credentials, as it continues to respond to a security incident that led to unauthorized access to application identities associated with Cloud Backup. The company said it has completed the following actions since it began
An attacker used a contributor's GitHub account to replace the download archives on five existing releases of OVHcloud's command-line tool on Monday, October 19. The account had been hijacked, and the new files were a build designed to steal secrets stored on the disk of any machine that runs it, the cloud provider says. They were live on GitHub for up to 94 minutes, from 12:47 to 14:21 UTC,
Microsoft urges organizations to secure AI Agents as insider threats, using dynamic permissions, continuous risk monitoring and automated containment. Microsoft published a detailed blog post explaining how organizations should secure AI agents. Microsoft sees them as a growing insider threat, not just a possible risk in the future. AI agents are not dangerous by nature, but they can access critical systems and may fail, be manipulated, or be compromised. “The most trustworthy agentic system will not be the one with the model we trust most. It will be the one designed that enables us to trust the model least.” said Satya Nadella This is not just a theory. It is a principle that should guide how organizations design and secure AI systems. The post makes a distinction worth holding onto. An agent that starts a task with the right permissions and behaves correctly can still become dangerous if its context changes mid-run. A prompt injection in an external document, a piece of untrusted data that shifts the agent’s reasoning, a task that evolves in a direction nobody anticipated, none of these require the agent to be compromised in the traditional sense. The model produces variable outputs, the context changes, and the agent’s behavior drifts. At machine speed, the window between that drift starting and it causing damage is very short. Microsoft’s proposed answer is continuous containment rather than upfront trust. You define what the agent is allowed to do before it acts, you monitor what it actually does while it runs, and you adjust its permissions in real time as its behavior or the risk calculation changes. An expense approval agent should start with access scoped exactly to reviewing and approving expenses, nothing broader, and those permissions should shrink or disappear the moment the agent tries something outside that task. The post introduces Dynamic Permissions to apply the principle of least privilege to AI agents. An agent should not automatically get the same access rights as the employee who deployed it. It starts with no permissions and receives only the access needed for its approved task. These permissions remain active only while the task is running. “The employee retains the employee’s access.” Microsoft states. “The agent receives its own identity, permissions and its own authority scoped to its current task and for the duration of the task only.” Alongside that, Dynamic Agent Risk is a continuously updated score that combines signals from identity, endpoints, network, cloud, runtime, and the agent’s own behavior to assess whether what the agent is currently doing still matches what it was authorized to do. The response when risk rises is graduated rather than binary: preserve reads while blocking writes, narrow accessible resources, require human approval for high-impact steps, or stop execution entirely if the risk warrants it. “Dynamic Agent Risk combines signals from identity, data, endpoints, cloud, network, runtime, and the agent’s own activity. It considers how securely the agent is configured, what resources it can reach, the potential impact if something goes wrong, and whether its actions are drifting from the user’s or organization’s intent.” Microsoft states. “The result is not a one-time score assigned at onboarding. It is a continuously updated understanding of the agent’s current risk and whether its current authority still matches that risk. “ Organizations frequently have AI agents running locally or introduced through third-party tools that were never formally registered in any central system. Microsoft says it’s expanding discovery to 60 types of local agents, including Claude Code, OpenAI Codex, and GitHub Copilot. Getting those into a unified registry with accountable identities is the prerequisite for applying any of the governance controls described in the rest of the post. You can’t set least-privilege permissions for an agent you don’t know exists. You can’t detect behavioral drift from an agent that doesn’t have a tracked identity. The agent registration problem is logistically harder than it sounds, especially for organizations where individual employees can spin up agents without IT involvement, which is most organizations right now. Microsoft recognizes that organizations cannot manually review every risk alert when they run hundreds of thousands of AI agents. Its solution is Guardian Agents, AI agents focused on security. They follow playbooks created by human operators, investigate changes in risk, apply or recommend security measures, and use what they learn to improve security policies. Humans set the goals and approve important decisions, while the agents handle continuous monitoring and response. However, this approach raises another concern. Guardian Agents are also AI agents, so they can fail or be manipulated, just like the agents they are supposed to control. Microsoft says security controls must operate outside the AI model and must not be changed by it. But this raises an important question: how can organizations ensure these controls remain reliable when the system enforcing them is also an AI agent? This is a challenge that real-world deployments will need to address in the coming years. “Organizations are adopting agents to serve customers better, solve problems faster, and give people more time for the work that matters most. The goal is not to make every model perfectly predictable before it can be useful.” Microsoft concludes. “The goal is to build systems that remain secure even when the model is not.” Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, AI Agents)
Confirm this action.
Leaving now will discard your changes.