Red Teaming Stories

Asian Stories

Red Teaming stories

Autonomous AI agents hit Asian government in four-day breach

Up to 85 government accounts were compromised in a four-day campaign that also reached nuclear and energy organisations, researchers said.

By Joseph Gabriel Lagonsin • 4 min read • Yesterday

Data Protection

ITSEC launches cybersecurity & AI academy in Jakarta

Indonesia's digital skills gap is fuelling demand for practical cyber and AI training, as firms struggle to hire workers.

By Sofiah Nichole Salivio • 5 min read • 2 days ago

Global Stories

Red Teaming stories

Security Information and Event Management

When AI memory, simulation and provenance collide

Realistic-looking security evidence can be fabricated when memory, simulation and model hallucinations blur provenance in AI workflows.

By Arjun • 5 min read • 3 days ago

Risk & Compliance

OpenAI expands Daybreak with specialist cyber model

Approved defenders will gain broader access to cyber tools as the new tiered Daybreak programme adds GPT-5.6-Cyber for advanced security work.

By Mark Tarre • 4 min read • 3 days ago

Encryption

OpenAI says Astra may hit critical cyber threshold

Tighter safeguards and pauses on internal work follow early tests that left OpenAI unable to rule out critical cyber capabilities in Astra.

By Mark Tarre • 4 min read • 4 days ago

Digital Transformation

CrowdStrike launches AI security challenge with AWS

Defenders will test prompt-injection tactics against AI agents as CrowdStrike and AWS offer USD $100,000 in prizes for a virtual red-team contest.

By Sean Mitchell • 4 min read • Last week

Robotics

Linux Foundation seeks AI security incident sharing rules

A proposed framework could help firms share AI security failures confidentially, as concern grows over agentic systems that act across corporate tools.

By Mark Tarre • 5 min read • Last week

Software-as-a-Service

Istari launches AI security managed operations with Cranium

Enterprises can now outsource day-to-day AI governance and monitoring, as the service aims to close a growing gap in live system oversight.

By Joseph Gabriel Lagonsin • 4 min read • Last week

Software-as-a-Service

Google adds security controls for AI agents in Chrome

Browser-based AI agents will face tighter enterprise oversight as Google moves to curb data leaks and unauthorised actions in Chrome.

By Mark Tarre • 4 min read • Last week

Hybrid Cloud

BeyondTrust warns identity risks drive most attacks

Three out of four offensive security investigations traced back to identity or privilege, exposing access gaps across cloud, SaaS and AI systems.

By Joseph Gabriel Lagonsin • 4 min read • Last week

Software-as-a-Service

Horizon3 raises USD $250 million at over USD $2 billion

The new capital will fund Horizon3's expansion into Asia-Pacific and Europe as demand for its attack-testing software grows among large enterprises.

By Joseph Gabriel Lagonsin • 4 min read • Last week

Network Security

Anthropic says Claude models accessed real systems in tests

Misconfigured test setups let three Claude models touch live systems, exposing production data and credentials during security exercises.

By Mark Tarre • 4 min read • Last week

Data Protection

EU AI Act deadline exposes growing agentic security risks

Security teams face fresh pressure as test breaches show autonomous AI can stray beyond scope, just as the EU AI Act tightens across Europe.

By Sofiah Nichole Salivio • 6 min read • Last week

Cloud Security

Acalvio launches AI deception guardrails for agents

Security teams may get earlier warning of hijacked AI agents as fake credentials and decoy tools are planted to expose misuse.

By Joseph Gabriel Lagonsin • 4 min read • Last month

Stories from Other Regions

Red Teaming stories

Privileged Access Management

AI agent used fake identities to push malicious code

Cybersecurity experts warn single-person approvals are now vulnerable after an AI agent used fabricated identities to slip malicious code past checks.

By Sean Mitchell • 6 min read • 2 days ago

Phishing

UK AI tests find Claude & GPT-5.6-Sol rogue actions

The findings heighten concern that frontier AI agents can breach boundaries, pressure real people and target software supply chains under loose controls.

By Joseph Gabriel Lagonsin • 5 min read • Last week