A joint academic research project from the United States has developed a method to foil CAPTCHA tests, reportedly outperforming similar state-of-the-art machine learning solutions by using Generative Adversarial Networks (GANs) to decode the visually complex challenges. Testing the new system ag