AI + Cybersecurity
Research and practical guidance covering AI-assisted threats, prompt injection, LLM security, deepfake risk, AI governance and defensive use of artificial intelligence.
Focused on responsible, defensive and security-conscious use of AI technologies.
Browse by Topic
Explore AI security research areas across LLM risk, governance and defensive use cases.
Enterprise LLM risks, data leakage controls and secure deployment guidance.
ExploreInput validation, output filtering and architectural defences for LLM-integrated applications.
ExploreResearch into AI-assisted phishing, impersonation and social engineering techniques.
ExploreSynthetic media risks, identity verification challenges and defensive awareness.
ExplorePolicy, risk assessment and oversight models for responsible AI adoption.
ExploreResponsible use of AI for detection, triage, investigation and security operations.
ExplorePublished AI Security Content
Articles, tutorials and cyber labs focused on AI security, LLM risk and responsible defensive AI.
Focused on responsible, defensive and security-conscious use of AI technologies.
Organizations can reduce Microsoft 365 phishing and account-compromise risk by prioritizing phishing-resistant MFA, tuning Defender for Office 365 anti-phishing protections, addressing spoofing paths, and using documented threat intelligence capabilities during investigations.
Sep 17, 2026
AI is accelerating vulnerability discovery, exploit development and cyber defence. Explore how AI-assisted attacks are changing the vulnerability lifecycle in 2026 and what defenders can do to reduce their exposure.
Aug 25, 2026
Deepfake fraud hit $1.1 billion in US losses in 2025 alone. Here's the real attack mechanics behind cases like Arup — and what actually stops them.
Aug 25, 2026
Prompt injection is the #1 LLM risk. Learn how direct and indirect attacks work and the layered defenses that actually stop them.
Aug 25, 2026
AI is moving beyond phishing assistance into more operational roles in cyberattacks. This analysis examines how threat actors are using AI in 2026, what the evidence actually shows, and what security teams should prioritise.
Aug 25, 2026