{"type": "article", "title": "LLMs respond differently to harmful prompts when AI watermarking is used", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/llms-respond-differently-to-harmful-prompts-when-ai-watermarking-is-used", "original_source": "https://arstechnica.com/security/2026/09/ai-text-watermarking-can-make-models-more-vulnerable-to-adversarial-prompts/", "published": "2026-09-17T18:33:13+00:00", "accessed": "2026-09-17", "id": "llms-respond-differently-to-harmful-prompts-when-ai-watermarking-is-used"}