As an essential branch of the Internet of Things (IoT), the Internet of Vehicles (IoV) provides users with efficiency and convenience for travel and is gradually replacing vehicle ad hoc network (VANET) as an integral component of intelligent transportation systems. However, due to the rapidly changing network topology and various communication capabilities, the IoV is confronted with more complex network security risks than IoT, such as trust and reputation. To prevent dishonest or malicious nodes from interfering with the IoV communication, we have proposed a Gated Linear Unit (GLU) based trust management system (GTMS) with blockchain in this paper. In the GTMS, the trust level of the node is dynamically adjusted to each message sent, which utilizes the GLU network model with hybrid trust feature extraction to calculate the value instead of a fixed formula. In addition, we design a method based on the blockchain for storing the global trust value. Road Side Units (RSUs) record the trust level adjustment with the modified blockchain technology, which customizes mining difficulty according to the node’s condition. The experimental results demonstrate that the proposed GTMS can detect malicious nodes among the road simulation with greater accuracy than the state-of-the-art method.