What We Missed: Google Gemini Joins the AI Escape Party
Prefilling the Reasoning Channel: Output-Prefix Attacks on Reasoning LLMs
Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures
AEGIS: Audio Endogenous Guarding via Internal Signals Against Large Audio-Language Model Jailbreaks
Research on Models Engaging in Genie-Like Behavior
Decoding Guardrails: XAI-Guided Perturbation Analysis of Prompt Injection Detection
The Target Is No Longer the Model. It’s the Agent.
9 min
CASCADE Against Jailbreaks: Combination Across Stages with Controlled Attack-Defense Evaluation
HE-Guardrail: A Homomorphic Guardrail Against Jailbreak Attacks for Encrypted Large Language Model Inference
One runaway AI agent racked up a $50,000 cloud bill
in the wild 4 min3
Uncensored AI sold on hacking forum as alternative to ChatGPT and Claude jailbreaks
2 min
Divide, Consult, Conquer: Capability Laundering Through Aligned LLMs
New AI Attack Hides Malicious Instructions in Normal-Looking Text to Evade Safety Filters
5 min
PuzzleMask: Abusing Plain Prose as a Covert AI Attack Vector
15 min
Six Chinese AI firms accused of aggressively copying US frontier models
in the wild 8 min1
CS-Guard: Benchmarking LLM Guardrails for Code Generation Security
1
Stealing AI Reasoning Traces
2 min
Structural Jailbreaks Generalize but Do Not Compound: A cross-provider and multilingual study of Involuntary In-Context Learning
1
OpenAI's rebel agent swarm died young, but its chilling logs live on
5 min
Zero-Click Grok Chat History Theft: Adversa AI Demonstrates Cryptographic Context Injection
5 min
Bad Likert Judge: A Novel Multi-Turn Technique to Jailbreak LLMs by Misusing Their Evaluation Capability
15 min