Here's your daily roundup of the most relevant AI and ML news for January 19, 2026. Today's digest includes 1 security-focused story. We're also covering 7 research developments. Click through to read the full articles from our curated sources.
Security & Safety
1. Google Gemini Prompt Injection Flaw Exposed Private Calendar Data via Malicious Invites
Cybersecurity researchers have disclosed details of a security flaw that leverages indirect prompt injection targeting Google Gemini as a way to bypass authorization guardrails and use Google Calendar as a data extraction mechanism. The vulnerability, Miggo Security's Head of Research, Liad Eliya...
Source: The Hacker News (Security) | 6 hours ago
Research & Papers
2. Adversarial Poetry as a Universal Single-Turn Jailbreak Mechanism in Large Language Models
arXiv:2511.15304v3 Announce Type: replace-cross Abstract: We present evidence that adversarial poetry functions as a universal single-turn jailbreak technique for Large Language Models (LLMs). Across 25 frontier proprietary and open-weight models, curated poetic prompts yielded high attack-succe...
Source: arXiv - AI | 18 hours ago
3. From Adversarial Poetry to Adversarial Tales: An Interpretability Research Agenda
arXiv:2601.08837v2 Announce Type: replace-cross Abstract: Safety mechanisms in LLMs remain vulnerable to attacks that reframe harmful requests through culturally coded structures. We introduce Adversarial Tales, a jailbreak technique that embeds harmful content within cyberpunk narratives and pr...
Source: arXiv - AI | 18 hours ago
4. A Natural Primal-Dual Hybrid Gradient Method for Adversarial Neural Network Training on Solving Partial Differential Equations
arXiv:2411.06278v3 Announce Type: replace-cross Abstract: We propose a scalable preconditioned primal-dual hybrid gradient algorithm for solving partial differential equations (PDEs). We multiply the PDE with a dual test function to obtain an inf-sup problem whose loss functional involves lower-...
Source: arXiv - Machine Learning | 18 hours ago
5. Robust and Efficient Zeroth-Order LLM Fine-Tuning via Adaptive Bayesian Subspace Optimizer
arXiv:2601.01452v4 Announce Type: replace-cross Abstract: Fine-tuning large language models (LLMs) with zeroth-order (ZO) optimization reduces memory by approximating gradients through function evaluations. However, existing methods essentially perform updates in a one-dimensional space, and suf...
Source: arXiv - AI | 18 hours ago
6. Reinforcement Fine-Tuning for Materials Design
arXiv:2504.02367v3 Announce Type: replace-cross Abstract: Reinforcement fine-tuning played an instrumental role in enhancing the instruction-following and reasoning abilities of large language models. In this work, we employ reinforcement fine-tuning for materials design, in which discriminative...
Source: arXiv - Machine Learning | 18 hours ago
7. Exploring LLM Features in Predictive Process Monitoring for Small-Scale Event-Logs
arXiv:2601.11468v1 Announce Type: new Abstract: Predictive Process Monitoring is a branch of process mining that aims to predict the outcome of an ongoing process. Recently, it leveraged machine-and-deep learning architectures. In this paper, we extend our prior LLM-based Predictive Process Moni...
Source: arXiv - AI | 18 hours ago
8. Causal-SAM-LLM: Large Language Models as Causal Reasoners for Robust Medical Segmentation
arXiv:2507.03585v2 Announce Type: replace-cross Abstract: The clinical utility of deep learning models for medical image segmentation is severely constrained by their inability to generalize to unseen domains. This failure is often rooted in the models learning spurious correlations between anat...
Source: arXiv - AI | 18 hours ago
About This Digest
This digest is automatically curated from leading AI and tech news sources, filtered for relevance to AI security and the ML ecosystem. Stories are scored and ranked based on their relevance to model security, supply chain safety, and the broader AI landscape.
Want to see how your favorite models score on security? Check our model dashboard for trust scores on the top 500 HuggingFace models.