Red Teaming stories

UK cyber agency warns that misunderstanding AI prompt injection risks could trigger data breaches surpassing past SQL attacks in scale.

AI supply chains will consolidate by 2026, with tight human oversight, integrated infrastructure and trust becoming key differentiators.

Hack The Box launches HTB AI Range, the world’s first controlled AI cyber range for testing autonomous security agents in live scenarios.

AI red teaming blends security tactics with safety goals to prevent exploits in chatbots, defending users from harm beyond classic cyber threats.

Despite 89% piloting AI, 78% neglect core data security, leading to breaches largely from known vulnerabilities, warns new Tenable research.

The UK Government's Cyber Security and Resilience Bill raises standards for over 900 suppliers, enhancing cyber defence across critical national infrastructure and supply chains.

Check Point, Lakera and the UK AI Security Institute launch b3, an open-source benchmark to test security of large language models in AI agents.

Australian firms are urged to adopt proactive cybersecurity strategies, using the PEST framework to anticipate threats and enhance resilience in a complex landscape.

Check Point and Lakera launch the open-source b3 benchmark to boost AI agents' LLM security against real-world cyber threats with threat snapshots.

CybExer has secured EUR €494,200 to develop an AI-driven cyber defence solution enhancing organisational readiness and adaptive threat response.

Practical DevSecOps has certified over 1,000 professionals in its rigorous AI security programme, addressing critical skills gaps in defending against evolving AI threats.

Bugcrowd’s latest CISO report reveals hardware vulnerabilities surged 88% and network risks doubled amid rising AI use and connected devices in 2024.

F5 will acquire CalypsoAI for USD $180 million to enhance AI security, integrating its platform to protect enterprises from emerging AI threats.

JFrog integrates partners like GitHub and Sonar to unify audit and compliance in AppTrust, streamlining governance across the AI-driven software lifecycle.

Radware warns that autonomous AI agents in businesses expand cyber risk, urging stricter security as these systems create complex, hard-to-track attack pathways.

Rapid7 launches Vector Command Advanced, a platform combining automated and human-led tests to enhance security and simplify compliance with PCI, ISO 27001, and NIST.

Bugcrowd and Pretera partner to offer continuous, scalable security testing across finance, healthcare and government sectors in the Benelux region.

CISO report reveals AI, supply chain, and insider threats prompt security leaders to adopt offensive strategies and continuous testing for better resilience.

Kindo.ai triples enterprise adoption, launches Deep Hat AI model, enhancing AI-native automation for security, DevOps and IT operations with zero customer churn.

Pangea launches AIDR platform to bolster security as generative AI use grows, tackling threats like prompt injection in enterprise settings.