A data race occurs in computer programming when two or more threads access the same shared data simultaneously, and at least one of the threads modifies that data. This situation can lead to unpredictable behavior and bugs, as the outcome depends on the timing of the threads' execution. To prevent data races, developers often use synchronization mechanisms like mutexes or semaphores. These tools help ensure that only one thread can access the shared data at a time, maintaining data integrity and consistency in multi-threaded applications.