Summary
- Cisco’s AI Push: Cisco used the RSAC 2025 Conference to show off new AI security tools, pitching faster and more accurate threat detection.
- Focus on Real-Time Response: The new tools aim to help security teams respond to attacks as they happen.
- Industry Position: Cisco wants to be seen as a leader in AI-powered security.
- Open Questions: Analysts say the real test will be how well the tools work in customers’ networks.
AI Takes Center Stage
At the RSAC 2025 Conference in San Francisco, Cisco announced a set of AI-driven security products. The biggest headline was an open-weight AI model built for security work, released by Cisco’s Foundation AI team, which the company said security teams can run and customize themselves.
Real-Time Threat Detection and Response
Cisco also added AI agents and assistants to its security products, including its XDR and Splunk tools, to help analysts investigate alerts and decide what to do next. The pitch is speed. Lots of it. A human analyst can only read so many alerts. Software can triage thousands.
Start your morning with the signal that matters.
Weekday mornings. Five minutes. What changed in security, why it matters.
By subscribing you agree to our Privacy Policy.
Free. Weekday mornings. 5 minutes or less.
Cisco’s Strategy
Cisco has been building its security business for years, and its 2024 purchase of Splunk gave it a large security analytics platform. The RSAC announcements tie those pieces together with AI. Rivals made similar pitches. Palo Alto Networks did. So did Microsoft and CrowdStrike.
What Analysts Say
Analysts welcomed the moves but say the proof will come in real deployments. Open models are useful for security teams that want control over their data, but they still need tuning and oversight. AI that makes a wrong call at machine speed can cause problems as fast as it solves them.
What Comes Next
Cisco’s RSAC announcements show how central AI has become to security products. Customers should test these tools against their own data before trusting them with important decisions. The direction is clear. Security vendors expect AI to do more of the routine investigation work, leaving people to handle the hard cases.
Start your morning with the signal that matters.
Weekday mornings. Five minutes. What changed in security, why it matters.
By subscribing you agree to our Privacy Policy.
Free. Weekday mornings. 5 minutes or less.
Newer CyberExperts coverage on this topic
This article still works as background. If you want the current picture, start with the freshest related coverage below and today's brief.
Citrix NetScaler CVE-2026-19490: The Auth Bypass That Went From PoC to Probes in a Day
CVE-2026-19490 is a CVSS 9.3 NetScaler Gateway/AAA auth bypass. A public PoC on Sept 2 was followed by live probes within a...
PaperCut NG/MF: The Print Server That Learned Remote Code Execution Again
CVE-2026-81578 and CVE-2026-82078 chain to pre-auth RCE on PaperCut Application Server. Confirm Emergency Patch Release 2, restrict admin exposure, and hunt like...
JFrog Artifactory CVE-2026-82329: Empty Join Key, Full Admin Token, Busy Scanners
CVE-2026-82329 lets attackers forge Artifactory admin tokens via a deterministic empty join key. Mass scanning peaked near 406,000 attempts; patch fixed builds...
Thursday’s brief: Java4Shell first, then nginx, F5 BIG-IP and Synology
The fastest way to catch up on what changed after this article was published.
Start your morning with the signal that matters.
Get the biggest cybersecurity developments, why they matter, and where to go deeper on CyberExperts.
By subscribing you agree to our Privacy Policy.
Free. Weekdays. Built for operators.