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, Thursday 2 October 2025 Retrospective Covers 1 October 2025 00:00 to 2 October 2025 00:00 UTC Web: https://keryn.dev/headlines/daily/2025-10-02/ TOP DEVELOPMENTS [1] Data breach at dealership software provider impacts 766k clients BleepingComputer · Incident · https://bleepingcomputer.com/news/security/data-breach-at-dealership-software-provider-impacts-766k-clients [2] Google Drive for desktop gets AI-powered ransomware detection BleepingComputer · Incident · https://bleepingcomputer.com/news/security/google-drive-for-desktop-gets-ai-powered-ransomware-detection [3] Understanding Adversarial Transfer: Why Representation-Space Attacks Fail Where Data-Space Attacks Succeed arXiv · Research · https://arxiv.org/abs/2510.01494 SCA [4] Semantics-Aligned, Curriculum-Driven, and Reasoning-Enhanced Vulnerability Repair Framework arXiv · Research · https://arxiv.org/abs/2510.01002 [5] Maven-Lockfile: High Integrity Rebuild of Past Java Releases arXiv · Research · https://arxiv.org/abs/2510.00730 AGENTIC SDLC [6] Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking Attacks arXiv · Research · https://arxiv.org/abs/2510.01359 TRUST [7] A Call to Action for a Secure-by-Design Generative AI Paradigm arXiv · Research · https://arxiv.org/abs/2510.00451 MODELS [8] Fine-Tuning Jailbreaks under Highly Constrained Black-Box Settings: A Three-Pronged Approach arXiv · Research · https://arxiv.org/abs/2510.01342 [9] Where Do Backdoors Live? A Component-Level Analysis of Backdoor Propagation in Speech Language Models arXiv · Research · https://arxiv.org/abs/2510.01157 [10] Large Reasoning Models Learn Better Alignment from Flawed Thinking arXiv · Research · https://arxiv.org/abs/2510.00938 AI SECURITY [11] WAInjectBench: Benchmarking Prompt Injection Detections for Web Agents arXiv · Research · https://arxiv.org/abs/2510.01354 [12] Eyes-on-Me: Scalable RAG Poisoning through Transferable Attention-Steering Attractors arXiv · Research · https://arxiv.org/abs/2510.00586 ADVISORIES IN THIS WINDOW 9 npm advisories · 6 critical · 1 high · 2 low 5 Packagist advisories · 1 high · 4 low 3 PyPI advisories · 1 critical · 2 moderate 2 Maven advisories · 2 moderate 1 crates.io advisory · 1 critical All advisories: https://keryn.dev/headlines/advisories/ Titles, links and short excerpts only; every item links to its source. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.