Cequence Security can continue evaluating and evolving the machine learning and the AI that they have because attackers are using more AI, making it faster to learn how to attack APIs and business logic. When Cequence Security invests in their AI, they will provide quicker protection that our customers need, and the way that attackers think can be translated into security measures on Cequence Security workflows and baseline. They need to improve faster in the AI environment because the possibility of attackers is growing, which is something that must be improved continuously as we need quicker releases since the market demands it. Regarding accuracy and reliability, the AI has a baselining that allows it to learn from attacks, but it needs to improve more because in the Brazilian market, we have a creative environment for attackers who are familiar with the business logic of our customers, making it easier for them to imitate legitimate users during attacks. Thus, they need to enhance behavior analysis to differentiate themselves from all other products on the market. Latency can be a significant issue in bigger environments, especially since the architecture requires uninterrupted traffic to ensure customers can finish transactions. The defender component managing inline traffic holds all traffic and has mechanisms to control latency, allowing fallback during high-risk situations, but the sensor used for mirroring traffic is outdated and does not work properly.
