AI Microsoft found an LLM attack in the wild could swing enterprise cloud contracts Microsoft calls it AI Recommendation Poisoning. The prompt engineer behind CiteMET tells us "remember" was never intended to be coercive. Phillip de Wet Feb 13, 2026 - 5 min read Photo by Raghavendra V. Konkathi on Unsplash Get the full story: Subscribe for free Join peers managing over $100 billion in annual IT spend and subscribe to unlock full access to The Stack’s analysis and events. Subscribe now Already a member? Sign in