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, Friday 13 March 2026 Retrospective Covers 12 March 2026 00:00 to 13 March 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-03-13/ TOP DEVELOPMENTS [1] Detecting and analyzing prompt abuse in AI tools Microsoft · Research · https://microsoft.com/en-us/security/blog/2026/03/12/detecting-analyzing-prompt-abuse-in-ai-tools [2] AI-generated Slopoly malware used in Interlock ransomware attack BleepingComputer · Incident · https://bleepingcomputer.com/news/security/ai-generated-slopoly-malware-used-in-interlock-ransomware-attack [3] Hunting CUDA Bugs at Scale with cuFuzz arXiv · Research · https://arxiv.org/abs/2603.12485 DAST [4] Human in the Loop for Fuzz Testing: Literature Review and the Road Ahead arXiv · Research · https://arxiv.org/abs/2603.13411 REMEDIATION [5] Veeam warns of critical flaws exposing backup servers to RCE attacks BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/veeam-warns-of-critical-flaws-exposing-backup-servers-to-rce-attacks [6] Apple Issues Security Updates for Older iOS Devices Targeted by Coruna WebKit Exploit The Hacker News · Analysis · https://thehackernews.com/2026/03/apple-issues-security-updates-for-older.html TRUST [7] iPhones and iPads Approved for NATO Classified Data Schneier on Security · Analysis · https://schneier.com/blog/archives/2026/03/iphones-and-ipads-approved-for-nato-classified-data.html MODELS [8] Accelerating Suffix Jailbreak attacks with Prefix-Shared KV-cache arXiv · Research · https://arxiv.org/abs/2603.13420 [9] Cascade: Composing Software-Hardware Attack Gadgets for Adversarial Threat Amplification in Compound AI Systems arXiv · Research · https://arxiv.org/abs/2603.12023 [10] HomeSafe-Bench: Evaluating Vision-Language Models on Unsafe Action Detection for Embodied Agents in Household Scenarios arXiv · Research · https://arxiv.org/abs/2603.11975 AI SECURITY [11] Security Considerations for Artificial Intelligence Agents arXiv · Research · https://arxiv.org/abs/2603.12230 [12] The Mirror Design Pattern: Strict Data Geometry over Model Scale for Prompt Injection Detection arXiv · Research · https://arxiv.org/abs/2603.11875 [13] OpenClaw PRISM: A Zero-Fork, Defense-in-Depth Runtime Security Layer for Tool-Augmented LLM Agents arXiv · Research · https://arxiv.org/abs/2603.11853 ADVISORIES IN THIS WINDOW 55 npm advisories · 25 critical · 7 high · 19 moderate · 4 low 20 NuGet advisories · 5 high · 15 moderate 11 PyPI advisories · 2 critical · 5 high · 2 moderate · 2 low 10 Go advisories · 1 critical · 5 high · 4 moderate 5 crates.io advisories · 2 high · 3 moderate 2 Multi-ecosystem advisories · 1 moderate · 1 low 1 Packagist 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.