Posts tagged #LLM Security

Generative AI

The Cracked Containment Vessel: How Agentic Sandbox Escapes are Rewiring Enterprise Trust

Frontier labs report critical AI sandbox escapes just as EU AI Act transparency mandates take effect...

Read more →

Threat Intelligence

Indirect Prompt Injection: How Hidden Web Instructions Are Hijacking AI Agents in Real-World Campaigns

Zscaler's ThreatLabz has uncovered two real-world campaigns utilizing indirect prompt injection to h...

Read more →