loader

Research

Our research teams investigate emerging threats, secure software, digital identity, and the societal impact of connected systems—so defenders can act on evidence, not fear.

Threat Research

Tracks adversaries, campaigns, exposure patterns, and the defensive signals organizations can use.

AI Security

Studies agent safety, model misuse, prompt attacks, and reliable controls for AI-enabled systems.

Privacy & Identity

Examines identity abuse, data exposure, verification, and privacy-preserving security systems.

Malware Analysis

Analyzes malicious files, infrastructure, and behavior to improve detection and response.

Vulnerability Research

Finds software weaknesses and develops practical, responsibly disclosed remediation guidance.

Human Solving Services and the Limits of Asking "Are You Human?"

Threat Research · Aug 25, 2026

A captcha solved by a paid human is solved by a human. The defence cannot be a harder puzzle, because the puzzle is not what failed. What actually separates a genuine visitor from a relayed one is continuity and cost.

Threat Research · Aug 27, 2026

What a Router Traffic Counter Actually Measures

A gateway counter is a total, not a rate. It resets without warning, and on a fast connection a 32-bit one can wrap in under a minute. Every honest traffic graph is built on top of those three facts, and the graphs that ignore them fail in the same three ways.

Vulnerability Research · Aug 27, 2026

A Monitoring Agent Is Not an API Client

Continuous telemetry and request-driven API calls arrive over the same transport and look identical at the door. Metering them the same way builds a system where a customer's own monitoring throttles their production integration — a denial of service nobody had to attack you to cause.

Malware Analysis · Aug 27, 2026

Antivirus and Anti-Cheat Want the Same Access. Only One of Them Is Trusted.

A resident security scanner needs kernel drivers, memory inspection and filesystem hooks. A kernel anti-cheat treats exactly those capabilities as the attack it exists to stop. On a competitive gaming machine, "just install antivirus" is not free advice.

Threat Research · Aug 22, 2026

An IP Flag Is Not a Fact

Every IP intelligence product answers "is this a VPN?" with true or false. The honest answer has a second half — how well-evidenced that is — and dropping it moves the risk onto whoever is blocking real users.

Publications

DateResearch teamTitle
Aug 27, 2026Threat ResearchWhat a Router Traffic Counter Actually Measures
Aug 27, 2026Vulnerability ResearchA Monitoring Agent Is Not an API Client
Aug 27, 2026Malware AnalysisAntivirus and Anti-Cheat Want the Same Access. Only One of Them Is Trusted.
Aug 25, 2026Threat ResearchHuman Solving Services and the Limits of Asking "Are You Human?"
Aug 22, 2026Threat ResearchAn IP Flag Is Not a Fact
Aug 22, 2026Threat ResearchWhy Residential Proxies Defeat the Standard Playbook
Aug 15, 2026Privacy & IdentityreCAPTCHA Alternatives: What Actually Differs
Aug 15, 2026Threat ResearchThe Controls Cyber Insurers Ask About
Aug 15, 2026AI SecurityPrompt Injection Is an Architecture Problem, Not a Filter Problem
Aug 15, 2026Privacy & IdentityWhy Credential Stuffing Survives MFA
Aug 2, 2026Threat ResearchDefending the Origin: Mitigating CDN Bypass and Multi-Vector Attacks