A security device may receive a request from a client device and intended for a server device. The security device may identify the request as being associated with a malicious activity. The malicious activity may include one or more undesirable tasks directed to the server device. The security device may generate a challenge-response test based on identifying the request as being associated with the malicious activity. The challenge-response test may be generated using one or more construction techniques. The security device may provide the challenge-response test to the client device. The security device may receive, from the client device, a proposed solution to the challenge-response test. The security device may identify the proposed solution as being generated using an optical character recognition (OCR) program. The security device may protect the server device from the client device based on identifying the solution as being generated using an OCR program.