This paper investigates the analysis and design of distributed fault detection for $\mathcal{L}_{2}$ observer-based intelligent connected vehicles in a zero-trust environment. Intelligent connected vehicles can improve traffic efficiency, but the system can be unreliable due to network security issues. Fault detection of intelligent connected vehicles is performed by introducing a zero-trust paradigm. First, the relative output estimation error is defined based on the communication topology of the intelligent connected vehicle. Then, a design scheme of $\mathcal{L}_{2}$ observer-based fault detection for the distributed system of intelligent connected vehicles in a zero-trust environment is proposed, and the stability and existence conditions of the observer for the distributed system of intelligent connected vehicles are verified. Finally, the feasibility and effectiveness of the design scheme are verified by simulation results.