.jpg)
DeepTempo, a pioneer in behavioral threat detection powered by deep learning for the identification of today's increasingly intelligent attacks, today announced its Intelligent Defense Platform, the first comprehensive cyber defense system built to deliver machine speed intelligence to enterprises, MSSPs, service providers, and critical infrastructure providers. With this move, DeepTempo has expanded from a LogLM foundation model provider to embrace a system-level approach.
LogLM performs approximately 279 billion calculations per sequence, pretrained on billions of logs.
Recent research shows 67.2% of exploited CVEs in 2026 have been zero-days; 82% of detections in 2025 were malware-free.
LogLM has achieved <1% false positives and <1% false negatives without adaptation in recent deployments.
Partners include Cribl, Snowflake, AWS, NVIDIA, and others.
Supports agentic intelligence pluggable through skills using enterprise licenses for OpenAI, Gemini, Claude, and on-premises reasoning models.
Proven at Deutsche Telekom, BNY, Stanford University, and in OT/ICS environments.
"When 82% of intrusions arrive without malware and breakout times are measured in seconds, you need a system to decide what actions to take and to capture end-to-end performance for continuous improvement," said Evan Powell, CEO and Founder, DeepTempo. "We built the Intelligent Defense Platform to augment what organizations already have, making every detection and workflow measurably better."
DeepTempo anticipated today's reality - the emerging use of AI by attackers is overwhelming human speed cyber security systems. While much has been made of the efficacy of Mythos and similar models in finding vulnerabilities, substantial evidence suggests that attackers are also using AI to orchestrate and execute campaigns that traditional systems struggle to identify and isolate.
The DeepTempo Intelligent Detection Platform extends existing cybersecurity investments by adding an intelligence layer across threat intelligence, detection, threat hunting, response, and related workflows, making every SIEM, SOAR, and AI SOC more focused and effective, reducing MTTD and MTTR while controlling spending on both human and AI intelligence.
DeepTempo's Intelligent Defense Platform builds upon the LogLM, which was pretrained on billions of logs and performs approximately 279 billion calculations per sequence. The LogLM uncovers complex, compound behavioral patterns that no human-authored rule can anticipate while eliminating the costly and error-prone retraining that undermines traditional anomaly detection.
Use cases in cyber traditionally have been fragmented across countless point products. Investments in telemetry platforms such as Cribl and data lakes such as Snowflake have made the Intelligent Defense Platform possible. Running in customer environments, DeepTempo's IDL delivers insights and, optionally, takes actions without the cost, delays, and risk of siloed products that backhaul telemetry into their SaaS solutions for analytics.
Pluggable Architecture Extends Capabilities while Reducing Lock-in: DeepTempo has partnered with Cribl, Snowflake, and others at the data layer, and works well with Splunk and other SIEMs, as well as agentic solutions within the SOC. Agentic intelligence is pluggable through skills and similar patterns, allowing users to leverage their own AI solutions.
End-to-End Validation and Monitoring: Continuously evaluates the efficacy of existing rule-based and ML-based detections alongside LogLM-generated detections and can also be used to measure the efficacy and projected costs of many workflows.
Broader Telemetry used by the LogLM: DeepTempo's LogLM has broadened its capabilities and can now ingest network flow, firewall, DNS, WAF, cloud performance, commonOT, and agentic AI logs. In some recent deployments, the LogLM has achieved <1% false positives and <1% false negatives without any adaptation required.
Edge-Appropriate Deployment: DeepTempo has simplified the deployment and management of LogLM and related software at the edge. Distilled versions of the LogLM run on small systems, adding the ability to see novel and rapidly evolving attacks to systems running in critical infrastructure.
This announcement follows DeepTempo's recent launch of Vigil, the first open-source AI SOC built on an LLM-native architecture, underscoring the company's commitment to providing security teams with a more transparent, extensible foundation for modern security operations.
About DeepTempo
DeepTempo builds the AI-native intelligent defense platform for cybersecurity, powered by LogLM - a patented, purpose-built foundation model that identifies attacker intent directly from structured telemetry without relying on signatures or brittle rules. Proven at Deutsche Telekom, BNY, Stanford University, and in OT/ICS environments with the Technology Advancement Center. Available on the Snowflake Native App Marketplace and for deployment across cloud, hybrid, and on-premises environments. DeepTempo is a proud partner of Cribl, AWS, NVIDIA, Snowflake, and other leading solution providers.
TechIntelPro empowers B2B technology decision-makers with the latest industry trends, deep insights, leadership perspectives and actionable intelligence to drive both organizational and professional milestones—and thrive on the tech pulse.