TensorFlow Object Detection for one type of object

Hello guys, I am new to this forum, particularly ML and TensorFlow. To start off with my first project, I chose to follow a guide on object detection(this is the link to the guide Object Detection Guide). The object I want to detect is an IC Chip, only. So I altered some parameters and codes inside the code to match my preference. Everything went well, I am able to detect IC Chips with 99% accuracy, BUT, when I inserted a random image of a cat, the AI detects the cat as IC Chip, as well as other images I inserted.  In my test and training directory, I only put images that are IC Chips, I did not insert any other images. Please advice what should I improve? Any help is much appreciated! Thanks in advance