Tag
LLM red teaming has gone from an esoteric activity to a mandatory practice. With the OWASP Agentic Top 10 and the CSA Agentic AI Red Teaming Guide converging on shared vocabulary, this is the operational playbook any team deploying agents needs to have.
In 2023, software supply chains became attackers' favourite target: MOVEit exposed data from hundreds of organisations through a zero-day flaw, 3CX shipped a trojanised installer to millions of users, and npm and PyPI kept receiving malicious typosquatted packages. The practical defence combines SBOM, artefact signing with Sigstore, SLSA maturity levels, and continuous dependency scanning.
The NIS2 Directive expands European cybersecurity from 7 to 18 sectors, mandates 10 minimum technical measures and 24-hour incident notification, and imposes fines of up to 10 million euros or 2% of global turnover, with personal liability for management bodies that fail to comply.
Adversarial machine learning studies deliberate attacks on AI systems (evasion, poisoning and model extraction) and the defenses used to resist them, chiefly adversarial training, robustness certification and monitoring the distribution of production input data.
Cybersecurity combines technical controls (encryption, MFA, network segmentation), team training and an incident response plan to reduce the risk of ransomware, phishing, malware and identity theft. No single measure is enough: effectiveness depends on applying every layer at once and auditing the system regularly.
