Andriy Onufriyenko/Moment via Getty Images Follow ZDNET: Add us as a preferred source on Google. ZDNET's key takeaways AI has created new cybersecurity threats as well as solutions. Sentinel is now able to respond to threats more autonomously. Microsoft is capitalizing on the industry's period of transition. The rise of AI is reshaping cybersecurity, presenting both new threats and new tools for mitigating them. Microsoft has been seizing this moment of transformation, aiming to become cybersecurity teams' go-to resource in the burgeoning AI era. Also: Navigating AI-powered cyber threats in 2025: 4 expert security tips for businesses In its latest step toward that goal, the tech giant announced Tuesday that it has upgraded Sentinel -- a Security Management and Events Incidents (SIEM) platform designed to help cybersecurity professionals track and respond to threats. The platform is now more agentic, or capable of taking action autonomously without meticulous human oversight. The upgrades (and what they do) The first upgrade announced Tuesday has to do with the approach that Sentinel takes in its response to cyberthreats. According to Microsoft, the platform now operates according to what's known in the cybersecurity industry as "graph-based" context. Just as any point on a Cartesian graph can be plotted with precise coordinates along multiple axes, a graph-based cybersecurity response system breaks a computer network down into a unified system of interconnected digital pathways, so that a threat that pops up in any particular node can be assessed according to its relationship and interconnectivity with all the others. Also: How AI-driven automation is the key to unlocking your operational resilience The pathways a threat takes throughout a given system, by extension, can be plotted into the past or the future. These graph-based abilities are specifically designed for Security Copilot agents, which Microsoft debuted in March. "Building on Sentinel's graph-based context, Security Copilot agents can now reason more effectively across your environment -- correlating alerts, enriching context with relationships, prioritizing by impact, and automating common actions," Vasu Jakkal, Corporate Vice President of Microsoft Security, wrote in a company blog post. As is often the case with agentic AI systems, Jakkal positioned the newly upgraded Sentinel as a quick-fix solution for workers looking to automate routine workplace tasks so they can turn their attention to more fulfilling and challenging aspects of their jobs. Also: 96% of IT pros say AI agents are a security risk, but they're deploying them anyway "Work shifts from manual triage to agent-led workflows: agents orchestrate and automate routine tasks, while analysts review and approve outcomes -- focusing their time on strategic decisions and proactive threat hunts," he said. Microsoft also debuted the Sentinel Model Context Protocol (MCP) server, which allows custom agents built on Microsoft's Visual Studio (VS) Code or other code-editing platforms to be integrated into Sentinel. The company added that its Sentinel data lake -- an internal repository for the storage of structured and unstructured data, first launched in preview in July -- is now generally available. Bottom line Microsoft is framing the newly upgraded Sentinel as more than just another AI product for cybersecurity teams: it's being portrayed effectively as part of a broader paradigm shift for the industry, one that can be characterized by an active and ongoing collaboration between humans and AI. Also: Why AI-powered security tools are your secret weapon against tomorrow's attacks "The advances announced today are the building blocks for a new generation of defense," Jakkal writes of the new upgrades to Sentinel. Other tech developers have been stepping up to help businesses navigate the novel data security and governance challenges posed by agents and other emerging AI tools. Meta, for example, recently teamed up with software company CrowdStrike to develop a set of benchmarks that enterprises can use to test the capabilities of various AI-powered cybersecurity tools.