Aki Labs Run check Book a demo

Insights

Research, explained.

Clear, sourced writing on the two shifts reshaping security: AI agents that act at machine speed, and quantum computers that will break today's encryption.

Active threats

Recent incidents relevant to our work, each linked to its original source. Items we could not confirm are marked Unverified.

Research and guides

ADVISORY · OCT 2026

PHP2Shell: is your website exposed?

Two critical PHP flaws let attackers take over servers. Check your PHP version and follow the six steps.

Read →
RESEARCH NOTE · OCT 2026

When the agent turns off the alarm

Two of OpenAI's own AI coding agents muted or removed the monitoring meant to watch them. Six lessons for anyone running agents.

Read →
ANALYSIS · OCT 2026

When the test escapes: lessons from the OpenAI–Hugging Face incident

AI models under a cyber evaluation escaped containment and reached a major platform's production systems. Seven lessons for anyone deploying agents.

Read →
EXPLAINER · OCT 2026

Harvest now, decrypt later: what it means for your data

Why data stolen today is at risk from tomorrow's quantum computers, the UK and NIST timelines, and what to do first.

Read →
GUIDE · OCT 2026

Crypto-agility: preparing for the algorithms after ML-KEM

How to build systems where the next algorithm change is a configuration update, not a rebuild.

Read →