Cisco Extends Security Reach to AI Agents
None
<p>Cisco today at the <a href="https://www.rsaconference.com/usa">RSA Conference</a> (RSAC) extended its cybersecurity portfolio to <a href="https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2026/m03/cisco-reimagines-security-for-the-agentic-workforce.html">secure artificial intelligence (AI) agents</a> while at the same time employing AI to automate security operations.</p><p>At the core of that effort are extensions to the Cisco Duo identity and access management (IAM) platform that make it possible to discover them and apply security policies via a Model Context Protocol (MCP) gateway and the Cisco Secure Access, a security service edge (SSE) platform. That capability will enable cybersecurity teams to assign specific fine-grained permissions that only allow AI agents to access specific resources for a limited amount of time.</p><p>At the same time, Cisco unveiled DefenseClaw, an open source secure agent framework that leverages the OpenShell runtime created by NVIDIA to securely automate the deployment of AI agents. The framework includes tools for scanning code created using AI skills tools and MCP integrations along with an AI bill of materials (AI BOM) and CodeGuard, a set of rules for embedding best security practices into an AI agent. DefenseClaw makes it possible for cybersecurity teams to both verify MCP servers and inventory every AI asset.</p><p>Additionally, Cisco is making available an Explorer Edition of Cisco AI Defense, a lower cost version of an existing platform that enables cybersecurity teams to test AI models and applications in a way that can be integrated with multiple continuous integration/continuous delivery (CI/CD) platforms and application development platforms. Cisco also launched an Agent Runtime Software Development Kit (SDK) that can be used to embed security policies into software builds.</p><p>Finally, Cisco is expanding its Splunk Enterprise Security for managing security operations by adding an ability to continually analyze risk scores and relationships between IT assets. There are now a set of specialized AI agents to automate a range of tasks, including incident response and the building of automations that can be used to remediate an issue. There is also now a federated search tool that makes it simpler for cybersecurity teams to correlate data across multiple IT environments.</p><p>Jeff Shultz, senior vice president for portfolio strategy at Cisco, said in general Cisco is trying to fill three AI gaps that are emerging around infrastructure, trust and telemetry data. While interest in AI agents is high, many organizations are hesitant to deploy AI agents in a production environment.</p><p>In fact, a <a href="https://www.cisco.com/c/dam/m/digital/usc/Security-Advisory-Board/Agentic_AI_Infographic.pdf">recent Cisco survey</a> finds that while 85% work for organizations that are experimenting with AI agents, only 5% have thus far deployed them in a production environment.</p><p>Nevertheless, AI agent adoption is expected to rapidly accelerate and there is little time to replicate the security that has been provided to protect humans to AI agents, noted Shultz.</p><p>Just about every cybersecurity team will eventually need to find a way to secure AI agents, but for now adoption is exceeding the pace at which they are able to respond to yet another emerging technology. Unfortunately, that also makes it probable there will soon be multiple cybersecurity incidents involving AI agents that, for example, are subject to prompt injection attacks that are relatively trivial to create and launch. While cybersecurity teams will need time and funding to put the proper controls in place, it’s also certain that their incident response capabilities are also about to be put to the ultimate test.</p><div class="spu-placeholder" style="display:none"></div><div class="addtoany_share_save_container addtoany_content addtoany_content_bottom"><div class="a2a_kit a2a_kit_size_20 addtoany_list" data-a2a-url="https://securityboulevard.com/2026/03/cisco-extends-security-reach-to-ai-agents/" data-a2a-title="Cisco Extends Security Reach to AI Agents"><a class="a2a_button_twitter" href="https://www.addtoany.com/add_to/twitter?linkurl=https%3A%2F%2Fsecurityboulevard.com%2F2026%2F03%2Fcisco-extends-security-reach-to-ai-agents%2F&linkname=Cisco%20Extends%20Security%20Reach%20to%20AI%20Agents" title="Twitter" rel="nofollow noopener" target="_blank"></a><a class="a2a_button_linkedin" href="https://www.addtoany.com/add_to/linkedin?linkurl=https%3A%2F%2Fsecurityboulevard.com%2F2026%2F03%2Fcisco-extends-security-reach-to-ai-agents%2F&linkname=Cisco%20Extends%20Security%20Reach%20to%20AI%20Agents" title="LinkedIn" rel="nofollow noopener" target="_blank"></a><a class="a2a_button_facebook" href="https://www.addtoany.com/add_to/facebook?linkurl=https%3A%2F%2Fsecurityboulevard.com%2F2026%2F03%2Fcisco-extends-security-reach-to-ai-agents%2F&linkname=Cisco%20Extends%20Security%20Reach%20to%20AI%20Agents" title="Facebook" rel="nofollow noopener" target="_blank"></a><a class="a2a_button_reddit" href="https://www.addtoany.com/add_to/reddit?linkurl=https%3A%2F%2Fsecurityboulevard.com%2F2026%2F03%2Fcisco-extends-security-reach-to-ai-agents%2F&linkname=Cisco%20Extends%20Security%20Reach%20to%20AI%20Agents" title="Reddit" rel="nofollow noopener" target="_blank"></a><a class="a2a_button_email" href="https://www.addtoany.com/add_to/email?linkurl=https%3A%2F%2Fsecurityboulevard.com%2F2026%2F03%2Fcisco-extends-security-reach-to-ai-agents%2F&linkname=Cisco%20Extends%20Security%20Reach%20to%20AI%20Agents" title="Email" rel="nofollow noopener" target="_blank"></a><a class="a2a_dd addtoany_share_save addtoany_share" href="https://www.addtoany.com/share"></a></div></div>