United StatesTechnology and telecom2025 – ongoingTrellix
Wise agentic AI for security alert analysis
Trellix's Wise agents saved about eight hours of analyst time for every 100 security alerts
Problem
Security teams face a flood of alerts every day. Attackers can move fast enough to take data before analysts notice an intrusion. Trellix wanted AI that triages events on its own instead of adding another tool that produces more alerts.
Approach
Trellix built Wise, an agentic system that runs on Claude through Amazon Bedrock. It analyzes events across endpoint, network, email and cloud, correlates them with threat intelligence and recommends a response. Trellix gave the agents broad security objectives and grounded them in documentation. It did not fine-tune the model.
Results
Takeaway
Trellix measured the gain in analyst hours per batch of alerts. That let it describe the value as extra staff capacity for a typical customer.
Categories
- Industry
- Technology and telecom
- Business function
- IT and R&D
- Innovation type
- Product and service innovation
- AI technology
- AI agentsGenerative AI
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Anthropic's customer story describes Trellix's Wise platform, which uses Claude in Amazon Bedrock to triage security alerts across several environments. It reports about eight hours of staff time saved per 100 alerts, equal to roughly 10 additional analysts for a typical customer. The page carries no date; the Internet Archive first saved it on September 20, 2025.
Full details are in the original sources.