Intezer and Torq Enable First Agent-to-Agent AI Collaboration for Smarter SOC Automation

Security operations teams are under pressure to do more with less - less time, fewer people, and a constant surge of alerts that demand attention. Intezer and Torq have partnered to bring agentic AI collaboration into live SOC environments to reduce the operational load and speed up response time.

Instead of working in silos, Intezer’s forensic-grade AI agents now integrate with Torq’s HyperSOC system and its AI SOC Analyst, Socrates. These specialized agents operate in a coordinated, autonomous loop, triaging, investigating, and remediating alerts while continuously sharing context and delegating tasks.

According to Itai Tevet, CEO and co-founder of Intezer, this model is a step beyond traditional AI SOC tools that function in isolation. “The biggest difference here is that customers will have best-of-breed agents for two different things. Intezer is very good at investigation and triage while Torq’s specialty is case management and remediation. It’s like having two experts on your team that specialize in two different things.”

This integration mirrors how elite security teams operate - only now, the collaboration is autonomous. Rather than automating a single step, the agents maintain a feedback-driven cycle in which alerts are analyzed, actions taken, and context updated without human intervention, unless escalation is necessary.

This setup is especially valuable in high-volume environments, where speed and accuracy must go hand in hand. Intezer’s agents rely on a layered approach, combining machine learning, large language models (LLMs), deterministic logic, and forensic analysis - to ensure precise, auditable decision-making. “We knew that if AI agents were going to operate independently, they had to make decisions that were both accurate and aligned with each customer’s unique environment.”

This precision powers more than just triage. Once an alert is analyzed, the reasoning and context are passed to Torq’s Socrates engine for autonomous remediation, ensuring a seamless handoff and full-cycle response.

By filtering and processing alerts with minimal manual input, the combined system helps security teams prioritize complex threats that still require human judgment. At a time when talent shortages and burnout make 24/7 monitoring increasingly difficult, this approach offers real relief.

It also hints at a broader shift in SOC operations: a move toward ecosystems of cooperating AI agents rather than all-in-one tools. This modular approach lets teams deploy specialized capabilities and orchestrate their interactions for smarter outcomes.

For MSSPs, where scale, customization, and multi-tenancy are key, the model is especially relevant. “We built our platform to support large alert volumes and multi-tenant environments that are common to see in MSSPs,” says Tevet. “Torq is also optimized for MSSPs, with RBAC and tenant-aware automation, allowing us to deliver consistent, tailored service across client environments.”