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 35 of 2024 · Retrospective · Covers 20 August 2024 11:00 to 27 August 2024 11:00 UTC

Read on the web

Top developments

Google tags a tenth Chrome zero-day as exploited this year

BleepingComputer · Incident · Remediation

GitHub Enterprise Server vulnerable to critical auth bypass flaw

BleepingComputer · Analysis · Remediation

Google AI Studio: LLM-Powered Data Exfiltration Hits Again! Quickly Fixed.

Embrace The Red · Research · AI Security

Researchers Identify Over 20 Supply Chain Vulnerabilities in MLOps Platforms

The Hacker News · Research · SCA

Hardware Backdoor Discovered in RFID Cards Used in Hotels and Offices Worldwide

The Hacker News · Research · Models

SAST

CrossInspector: A Static Analysis Approach for Cross-Contract Vulnerability Detection

arXiv · Research · SAST, SCA

Project Safe Source: Identifying potential vulnerabilities in Wolfi upstream

Chainguard · Analysis · Remediation, SAST, SCA

SCA

Trust, but Verify: Evaluating Developer Behavior in Mitigating Security Vulnerabilities in Open-Source Software Projects

arXiv · Research · SCA

DAST

CIPHER: Cybersecurity Intelligent Penetration-testing Helper for Ethical Researcher

arXiv · Research · DAST

The Facts About Continuous Penetration Testing and Why It's Important

The Hacker News · Analysis · DAST

Agentic SDLC

Microsoft Copilot: From Prompt Injection to Exfiltration of Personal Information

Embrace The Red · Analysis · Agentic SDLC, AI Security, Remediation

Top companies ground Microsoft Copilot over data governance concerns

Simon Willison · Analysis · Agentic SDLC, AI Security

Microsoft Patches Critical Copilot Studio Vulnerability Exposing Sensitive Data

The Hacker News · Research · Agentic SDLC, Remediation

Security Attacks on LLM-based Code Completion Tools

arXiv · Research · Agentic SDLC, Models

Remediation

SonicWall warns of critical access control flaw in SonicOS

BleepingComputer · Analysis · Remediation

CISA Urges Federal Agencies to Patch Versa Director Vulnerability by September

The Hacker News · Analysis · Remediation

“Something has gone seriously wrong,” dual-boot systems warn after Microsoft update

Ars Technica: Security · Analysis · Remediation

From Chaos to Consistency: The Role of CSAF in Streamlining Security Advisories

arXiv · Research · Remediation

Trust

Do Responsible AI Artifacts Advance Stakeholder Goals? Four Key Barriers Perceived by Legal and Civil Stakeholders

arXiv · Research · Trust

An Open Knowledge Graph-Based Approach for Mapping Concepts and Requirements between the EU AI Act and International Standards

arXiv · Research · Trust

Leveraging LLMs for the Quality Assurance of Software Requirements

arXiv · Research · Trust

Models

Advancing Adversarial Suffix Transfer Learning on Aligned Large Language Models

arXiv · Research · Models

Atoxia: Red-teaming Large Language Models with Target Toxic Answers

arXiv · Research · AI Security, Models

Celtibero: Robust Layered Aggregation for Federated Learning

arXiv · Research · Models

HTS-Attack: Heuristic Token Search for Jailbreaking Text-to-Image Models

arXiv · Research · Models

AI Security

Data Exfiltration from Slack AI via indirect prompt injection

Simon Willison · Analysis · AI Security

The dangers of AI agents unfurling hyperlinks and what to do about it

Simon Willison · Analysis · AI Security

Hide Your Malicious Goal Into Benign Narratives: Jailbreak Large Language Models through Carrier Articles

arXiv · Research · AI Security, Models

Advisories in this window

  • 33 npm advisories · 27 critical · 2 high · 3 moderate · 1 low
  • 19 Go advisories · 1 critical · 5 high · 13 moderate
  • 14 PyPI advisories · 1 critical · 1 high · 12 moderate
  • 5 Maven advisories · 4 high · 1 moderate
  • 2 Packagist advisories · 1 high · 1 moderate
  • 2 RubyGems advisories · 1 high · 1 moderate
  • 2 crates.io advisories · 1 high · 1 low
  • 2 Multi-ecosystem advisories · 1 high · 1 moderate
  • 2 NuGet advisories · 2 moderate

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.