R1, the latest large language model (LLM) from Chinese startup DeepSeek, is under fire for multiple security weaknesses. The company’s spotlight on the performance of its reasoning LLM has also ...