AI Models Overthink Problems—and It’s a Security Risk

IEEE Spectrum AI · 1 week ago

Researchers from Zhejiang University and Alibaba developed an evolutionary algorithm that generates logically inconsistent prompts to cause reasoning models to produce excessive internal monologues, creating a denial-of-service vulnerability. The attack produced outputs up to 26.1 times longer than normal responses on the MATH dataset and proved effective against models including DeepSeek-R1, Qwen3-Thinking, GPT-o3, and Gemini 2.5 Flash. The findings highlight a shared security weakness in modern reasoning-capable LLMs that providers should address through mitigation strategies.