< ciso
brief />
Tag Banner

All news with #detection engineering tag

151 articles

How Spanner Graph Powers Google Ads Fraud Detection

🛡️ Google's Ads team has moved its graph-based abuse detection platform onto Spanner Graph, replacing slow offline MapReduce batch jobs. The platform now runs multi-hop traversals on live production data with sub-second to single-digit millisecond latency. It spans a graph of more than 500 billion nodes and 1 trillion edges, helping Google uncover coordinated fraud rings in real time. The same architecture is available to enterprises in retail, finance, and security.
read more →

When Trusted AI Agents Resemble Cyber Attacks

🛡️ Enterprise AI agents often perform rapid, broad, and parallel actions that match patterns behavioral security systems flag as malicious. These accurate alerts—what the author calls benign true positives—require SOCs to determine intent by correlating agent ownership, task provenance, and authorization across identity, orchestration, cloud, and business systems. CISOs should prioritize agent identity and governance, investigation architecture, and team design to maintain visibility and avoid blind spots.
read more →

OpenAI Daybreak GPT Models Now on Amazon Bedrock

🔒 Eligible customers can access OpenAI’s specialized cybersecurity models Daybreak Red: GPT-5.6 Cyber and Daybreak Blue: GPT-5.6 Sol via the bedrock-runtime endpoint on Amazon Bedrock. Daybreak Blue supports defensive workflows like vulnerability discovery, detection engineering, and incident response, while Daybreak Red is intended for advanced, authorized tasks with stricter access controls. Models are available through US and Global cross-Region inference (CRIS) profiles; enrollment with OpenAI and AWS account approval are required, and inference data is not used for model training.
read more →

AI-driven threats reshape digital supply chain risk

🛡️ This article describes how frontier AI and autonomous agents are accelerating digital supply chain attacks, transforming adversary capabilities and shortening the time from compromise to data exfiltration. It highlights real-world supply chain incidents, emerging risks from vendor updates and shadow AI, and the need for unified, automated defensive frameworks. Organisations must adopt machine-speed detection and response to maintain resilience.
read more →

When Email Security Acts Like a Written Code

🛡️ The article compares traditional, signature-based email defenses to a codified legal system: fast and certain but blind to attacks that fall outside written rules. It argues that anomaly detection and behavioral models expand the code but retain the same limits when confronting novel compromise tactics. By contrast, a reasoning approach—like common-law courts—can weigh evidence and adapt to new cases, escalating high-stakes items to human decision-makers. The piece urges combining fast automated checks with human-in-the-loop adjudication for contested, high-risk messages.
read more →

Kaspersky SIEM correlation rules knowledge base

📘 Vendors often highlight the number of SIEM correlation rules and MITRE ATT&CK coverage, but sheer volume doesn’t guarantee effectiveness. A rule only matters if the SIEM actually receives the specific event types it targets, so deployment should start by verifying connected event sources, then selecting workable rules. Kaspersky has published a Knowledge Base that lists all SIEM rules, offers a Builder to match rules to existing telemetry, and a MITRE view showing coverage and gaps.
read more →

Quiet Week in Infosec: Focus on Fundamentals

📰 Cisco Talos summarizes a quieter infosec week emphasizing that foundational defensive work matters more than headlines. The post highlights new threats like the AI-integrated NIK C2 C2 framework and critical PHP vulnerabilities, and urges teams to know their assets, assign owners for exceptions, and maintain continuous behavioral monitoring. Practical detection advice and IOCs are included.
read more →

The Rise of the Threat Operator in AI-driven SOCs

🛡️ AI is beginning to take over many traditional security analyst tasks, from investigating alerts to executing response actions, prompting a reorganization of SOC work and decision-making. As AI handles operational workloads, the emerging Threat Operator role will direct AI-driven defense, translate business risk into priorities, and judge automated decisions. The shift requires new training, governance, and career paths that combine technical fundamentals with oversight and risk management.
read more →

From Federal SOCs to Talos: Practical Security Insights

🔎 Amy Ciminnisi interviews security researcher Ryan Fetterman about his path from federal agencies and consulting to research roles at Splunk and Talos. He describes hands-on detection engineering, early experiences with host-based analytics, and the evolving tactics of attackers. Ryan also discusses career-building strategies, the value of project-based learning, and avoiding burnout in cybersecurity.
read more →

Detecting DGA-generated domains in email spam

📧 Kaspersky describes how its email products prioritize the "From" field to identify spam and block known malicious domains. Spammers automate mass domain registration using domain generation algorithms (DGAs) that produce many similar domain names, evading static denylist approaches. Kaspersky's solution flags suspicious addresses, adds their domains to a "bad" database, and uses a cascade of classifiers to cluster lookalike domains and generate concise blocking rules deployed to corporate email protections. This approach enables blocking whole batches of DGA-based domains with a single rule to reduce spam, phishing, and malware risks.
read more →

Generative AI Erodes Email Behavioral Signals

🧭Proofpoint demonstrated at Protect 2026 how generative AI is making phishing and BEC messages fluent, personalized, and unique, undermining behavioral signals email filters used to rely on. Attackers can craft varied, plausible enquiries or hostile campaigns at scale, often using legitimate-looking senders and redirects to evade reputation checks. Proofpoint proposes intent-based detection layered with traditional methods and tiered AI analysis to identify what messages are trying to achieve rather than how they look.
read more →

Moving Past Autonomous Exploitation to Machine Speed

🔒 Frontier AI advances have accelerated exploitation well beyond defenders' remediation pace, creating an asymmetry where discoveries outstrip reliable fixes. The article argues that high-precision detection must be coupled with an integrated, autonomous remediation system combining multi-agent orchestration, context injection, zero-egress sandboxing and iterative validation. It outlines a 5-stage harness, three specialized agents, strict four-gate authorization, graduated autonomy and evidence-backed outputs to deliver reproducible, scalable machine-speed security.
read more →

Six Emerging Malware Trends Security Pros Should Track

🛡️ Malware is evolving rapidly as attackers leverage AI, supply-chain compromise, and social engineering to increase scale and sophistication. Infostealers, malicious packages in developer ecosystems, targeted ransomware, ClickFix social engineering, polymorphic malware, and AI-assisted techniques are highlighted as rising trends. Security teams must update detection, harden developer and CI/CD pipelines, and raise user awareness to counter these shifts.
read more →

Decision Models Reshape SOC Triage and Forensics

🛡️ In a month, decision-focused AI models like TypeSafe's Jev and OpenAI's Decisions API went from novel to widespread, offering typed choices and calibrated probabilities instead of text. These models suit SOC alert triage: they provide automatable probabilities, are cheap enough for every alert, and produce auditable, schema-bound outputs. However, calibration reflects the distribution of inputs, so missing context or stale environment knowledge can make confident answers wrong. The article argues the real advantage comes from investing in automated forensic evidence collection and organizational memory—process trees, live memory forensics, detonation results, and learned normal behavior—so models receive the right inputs. Measured calibration, continuous context refresh, and recording verdict outcomes are essential for safe automation.
read more →

GuardDuty adds five indicators to improve detections

🔒 Amazon GuardDuty Extended Threat Detection now attaches five new indicator types — MALICIOUS_PACKAGE, MISCONFIGURATION, REACHABILITY, SENSITIVE_DATA, and VULNERABILITY — to correlated attack-sequence findings. These indicators are sourced from Amazon Inspector, Amazon VPC Reachability Analyzer, Amazon Macie, and AWS Security Hub, and appear in the Indicators tab and GetFindings JSON. The added context helps promote weak runtime signals into higher-confidence CRITICAL findings for Amazon EC2, ECS, and EKS workloads, improving prioritization and reducing noise.
read more →

Start AI-Native Security Programs With Outcomes

🔒 Start with the security outcome the business needs most and map which people, money, time and technology constraints prevent the team from delivering that outcome consistently. Identify work that should run continuously—posture management, detection engineering, threat intelligence and alert investigation—and apply AI to automate evidence gathering, correlation and reassessment. Choose a single operational outcome, set guardrails and measure success against concrete business-focused metrics, not just activity counts.
read more →

CISA Endorses Cyber Decoys; FortiDeceptor 6.3

🛡️ CISA published guidance on Using Cyber Decoys to Strengthen Detection and Response, urging an assume-breach posture and use of honeypots, honeytokens, breadcrumbs, and tripwires. FortiDeceptor implements these concepts with decoy VMs, centralized lure management, and integration with the Fortinet Security Fabric. Version 6.3 expands decoy coverage to GitLab, IoT printers, and cloud connectors for S3, GitHub, and SharePoint, and supports automated response workflows.
read more →

Gartner’s ISOC: A New Layer for Security Ops

🔒 Gartner introduced the Integrated Security Operations Center (ISOC) category to address the limits of traditional SIEMs by unifying detection, investigation, case management and response. ISOC emphasizes native detection and response, centralized data ownership, persistent incident case objects and cross-domain correlation to reduce latency and operational friction. The aim is streamlined workflows and lower costs for lean security teams confronting AI-accelerated threats.
read more →

AWS Continuum Raises Bar for Autonomous Code Security

🔒 AWS reports that Continuum for code vulnerabilities achieved an 89.0% end-to-end pass rate on the CyberGym-E2E benchmark, passing 819 of 920 tasks within a 90-minute limit. The multi-agent system improved on prior public results across all measured stages (discovery, validation, remediation) and reached 93.7% when allowed to run longer. The evaluation was conducted under network-isolation and submission-review rules to ensure results came from analysis rather than retrieval.
read more →

Cortex XCOR: AI-Driven Autonomous Observability

🚀 This post introduces Cortex XCOR, an AI-native observability platform from Palo Alto Networks that aims to deliver autonomous root cause analysis and remediation. The platform centers on the XCOR Operator and specialized AI agents, including an AI SRE that autonomously investigates incidents and recommends fixes. XCOR integrates full-stack telemetry with cost optimization to balance visibility and spending.
read more →