An AI overthinking attack can tie a robot up for over a minute

Researchers at Michigan Technological University demonstrated an 'overthinking' attack that can disable robots relying on large vision-language models for over a minute by inserting text into the camera's field of view, causing the model to process the text indefinitely and halt decision-making.

Robots that read the world through cameras now lean on large vision-language models to interpret what they see and decide what to do next. These models handle images and text together, so any words that fall inside the camera frame become part of the input. A stop sign, a street name, a sticker on a wall. The overthinking behavior at the center of the attack Researchers at Michigan Technological University have shown that this reading … More https://www.helpnetsecurity.com/2026/07/15/robot-ai-overthinking-attack/ The post An AI overthinking attack can tie a robot up for over a minute https://www.helpnetsecurity.com/2026/07/15/robot-ai-overthinking-attack/ appeared first on Help Net Security https://www.helpnetsecurity.com .