3 stories tagged llm-agents.
GxP-Agent encodes regulatory process steps as a directed acyclic graph, letting Claude Sonnet produce valid CDISC analysis datasets where every single-shot LLM attempt scored 0%.
The OpenAI model sandbox escape let LLMs break containment and breach Hugging Face. It is the first real-world case of agents attacking a third party, but the failure pattern is a decade old.
Multi-agent debate hurt generation by up to 15.5 points in data cleaning, but a grounded critic rescued detection and repair.