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

Weekly, week 32 of 2024 · Retrospective · Covers 30 July 2024 11:00 to 6 August 2024 11:00 UTC

Read on the web

Top developments

Google fixes Android kernel zero-day exploited in targeted attacks

BleepingComputer · Incident · Remediation

Protect Your Copilots: Preventing Data Leaks in Copilot Studio

Embrace The Red · Incident · Agentic SDLC, AI Security

UK govt links 2021 Electoral Commission breach to Exchange server

BleepingComputer · Incident · Remediation

Extracting Prompts by Inverting LLM Outputs

Simon Willison · Research · AI Security, Models

Stressed Pungsan: DPRK-aligned threat actor leverages npm for initial access

Datadog · Research · SCA

SAST

Automated Software Vulnerability Static Code Analysis Using Generative Pre-Trained Transformer Models

arXiv · Research · SAST

SCA

StackExchange abused to spread malicious PyPi packages as answers

BleepingComputer · Analysis · SCA

Adversarial Robustness of Open-source Text Classification Models and Fine-Tuning Chains

arXiv · Research · Models, SCA

Elevating Software Trust: Unveiling and Quantifying the Risk Landscape

arXiv · Research · SCA

CoSAI Panel at DEF CON: Securing the Future of AI

CoSAI · Analysis · SCA

DAST

ARVO: Atlas of Reproducible Vulnerabilities for Open-Source Software

arXiv · Research · DAST

ShellFuzzer: Grammar-based Fuzzing of Shell Interpreters

arXiv · Research · DAST

Trust

First Analysis of the EU Artifical Intelligence Act: Towards a Global Standard for Trustworthy AI?

arXiv · Research · Trust

Unlocking the Potential of Binding Corporate Rules (BCRs) in Health Data Transfers

arXiv · Research · Trust

Configure GitHub Artifact Attestations for secure cloud-native delivery

GitHub · Analysis · SCA, Trust

Models

Scaling Trends for Data Poisoning in LLMs

arXiv · Research · Models

Mitigating Malicious Attacks in Federated Learning via Confidence-aware Defense

arXiv · Research · Models

Can Reinforcement Learning Unlock the Hidden Dangers in Aligned Large Language Models?

arXiv · Research · Models

Mission Impossible: A Statistical Perspective on Jailbreaking LLMs

arXiv · Research · Models

AI Security

WHITE PAPER: A Brief Exploration of Data Exfiltration using GCG Suffixes

arXiv · Research · AI Security

Advisories in this window

  • 49 npm advisories · 30 critical · 6 high · 9 moderate · 4 low
  • 44 Go advisories · 6 critical · 16 high · 21 moderate · 1 low
  • 17 Maven advisories · 4 critical · 8 high · 5 moderate
  • 9 Packagist advisories · 1 high · 7 moderate · 1 low
  • 6 PyPI advisories · 2 high · 3 moderate · 1 low
  • 2 RubyGems advisories · 2 moderate
  • 2 NuGet advisories · 2 moderate
  • 1 crates.io advisory · 1 low

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.