Currently free during beta - premium features coming soon. Subscribe now to lock in early access.
AI_SAFETY

EU Regulatory Changes

1550 changes tracked across 24 compliance frameworks including DORA, NIS2, GDPR, EU AI Act, Cyber Resilience Act, and more.

All DORA NIS2 GDPR CSRD MaRisk ISO27001 EU_AI_ACT CRA DSA DMA eIDAS2 SOC2 PCI_DSS HIPAA ISO42001 AMLD6 PSD3 DATA_ACT GPSR CER EUDR CVE BREACH AI_SAFETY
arXiv: Cross-Agent Campaign Attribution: Linking Asynchronous Attacks Across LLM Agents
This publication from arXiv, dated July 21, 2026, presents a new research paper on a novel security vulnerability termed "Cross-Agent Campaign Attribution." The paper demonstrates how attackers can...
Read analysis →
arXiv: GLID: Gated Local Intrinsic Dimension Repairs the Blind Spots of Face-Forgery Detectors
This publication introduces GLID, a new method for detecting AI-generated or manipulated facial images, known as deepfakes. The research identifies a critical weakness in current face-forgery detec...
Read analysis →
arXiv: Find Before You Fine-Tune: A Diagnostic Study of Small LLMs for Cybersecurity QA
This paper, published on arXiv in July 2026, presents a diagnostic study evaluating the performance of small large language models (LLMs) on cybersecurity question-answering tasks. It introduces a ...
Read analysis →
arXiv: When to Trust the Map: Confidence-Aware LLM Routing for Automotive CVE-to-ATM Mapping
This publication from July 2026 introduces a novel framework for using large language models to map automotive cybersecurity vulnerabilities (CVEs) to specific attack trees (ATMs) within the automo...
Read analysis →
arXiv: Broken Gates: Re-evaluating Web Bot Defenses in the Age of LLM Agents
This paper, published on arXiv, presents a critical security analysis of web bot defenses in the context of advanced Large Language Model (LLM) agents. The authors demonstrate that current state-of...
Read analysis →
arXiv: CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization
arXiv: RECEIPT: Deterministic, Reward-Hacking-Resistant Verification for White-Box Agentic XSS Discovery
arXiv: Attacking Graph Foundation Models Through Their Shared Representation
arXiv: CryptanalysisBench: Can LLMs do Cryptanalysis?
arXiv: PIP-NTT: Towards a Scalable Memory-Parallelized Accelerator for Iterative NTT in PQC
arXiv: Towards an Automated Test of LLM Security Knowledge
arXiv: Trusted Credentials, Untrusted Behavior: Benchmarking LLM-Agent Security in High-Performance Computing
arXiv: 0-Cyclic Equalizability of Binary Words Characterized by Hamming Weight
arXiv: ResearchArena: Evaluating Sabotage and Monitoring in Automated AI R&D
arXiv: SoK: Adversarial Robustness of the Variational Quantum Eigensolver via Red-Teaming
arXiv: They'll Verify. They Just Won't Act. How Authority Framing and Laundered Code Turn a Trusted Agentic CI/CD Pip...
arXiv: PTSan: A Practical Memory Safety Sanitizer for C/C++ with Pointer-Object Authority
arXiv: Sarus: Privacy-Preserving Multi-Vendor Perception Fusion via Homomorphic Encryption
arXiv: Code Division Modulation Layers Against Forgetting and Inference in Continual Gait Identification
arXiv: Exploiting Load/Store Leakage of Sparse Vectors for Key Recovery in HQC