.svg){kind=logo}
Articles by
The Year of the Agent: What Recent Attacks Revealed in Q4 2025 (and What It Means for 2026)
A breakdown of the real attacker techniques that shaped Q4 2025, from prompt leakage to indirect injection, and what these patterns mean for agent security in 2026.
California’s AI Laws Are About to Meet Reality
Starting January 2026, California will enforce new rules for customer-facing AI—shifting AI regulation from policy documents to live, in-production behavior.
Indirect Prompt Injection: The Hidden Threat Breaking Modern AI Systems
A deep dive into how Indirect Prompt Injection exploits the hidden ingestion surfaces of modern AI systems, why agentic models amplify the risk, and which system-level defenses actually work.
Agentic AI Threats: Over-Privileged Tools & Uncontrolled Browsing (Part 2)
This second installment in Lakera’s Agentic AI Threats series reveals how tool overreach and uncontrolled browsing create new exploit paths—and how runtime guardrails like Lakera Guard and Red keep agents safe.
2025 GenAI Security Readiness Report: A Clearer Picture of Where Enterprises Stand
Discover key insights from Lakera’s 2025 GenAI Security Readiness Report—adoption trends, top risks, and why confidence without vigilance isn’t readiness.
The Backbone Breaker Benchmark: Testing the Real Security of AI Agents
The Backbone Breaker Benchmark (b3) is the first human-grounded, threat-realistic benchmark for AI agents, making LLM security measurable, reproducible, and comparable across models.
Activate
untouchable mode.
untouchable mode.
Get started for free.
Lakera Guard protects your LLM applications from cybersecurity risks with a single line of code. Get started in minutes. Become stronger every day.
Join our Slack Community.
Several people are typing about AI/ML security. Come join us and 1000+ others in a chat that’s thoroughly SFW.
