PREVIEW. This is how it will look in an inbox; it was not sent to you. Each real email carries its own unsubscribe link and Keryn's postal address.

Keryn Headlines

Daily, Tuesday 18 February 2025 · Retrospective · Covers 17 February 2025 00:00 to 18 February 2025 00:00 UTC

Read on the web

Top developments

ChatGPT Operator: Prompt Injection Exploits & Defenses

Embrace The Red · Analysis · AI Security

Unveiling Privacy Risks in LLM Agent Memory

arXiv · Research · Models

BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model

arXiv · Research · Models

SAST

SmartLLM: Smart Contract Auditing using Custom Generative AI

arXiv · Research · SAST

DAST

A Survey of Fuzzing Open-Source Operating Systems

arXiv · Research · DAST

Models

ReVeil: Unconstrained Concealed Backdoor Attack on Deep Neural Networks using Machine Unlearning

arXiv · Research · Models

DELMAN: Dynamic Defense Against Large Language Model Jailbreaking with Model Editing

arXiv · Research · Models

Adversary-Aware DPO: Enhancing Safety Alignment in Vision Language Models via Adversarial Training

arXiv · Research · Models

Advisories in this window

  • 154 npm advisories · 154 critical

All advisories

Titles, links and short excerpts only; every item links to its source. Selection is automatic; the editor's note is AI-drafted and checked as described. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.