Hawk has announced the launch of a library of AI-powered fraud typology models, aimed at delivering personalised protection at speed.
Following this announcement, the newly introduced Fraud Day One Defense Models were developed in order to provide financial institutions with the needed access to personalised and efficient protection against common fraud attack vectors.
In addition, the models are expected to safeguard an institution’s customers from the beginning with the use of the Hawk fraud platform, aiming to defend them against typologies ranging from authorised push payment fraud and merchant fraud to money mule behavior and account takeover. The company is set to continue to focus on meeting the needs, preferences, and demands of clients and users in an ever-evolving market, while prioritising the process of remaining compliant with the regulatory requirements and laws of the industry as well.
More information on Hawk’s Fraud Day One Defense Models launch
According to the official press release, Hawk data scientists' tailored model was designed to meet the organisation’s need at speed, aiming to deliver trained models in less than three days after initial data and solution set up. This approach includes a highly automated feature selection process and model pipeline, which were developed to facilitate faster model tuning and result in precision prevention. At the same time, the process will give FIs the possibility to stop high-impact threats, distinct to their business from implementation, as well as effectively stop financial crime in its tracks.
Furthermore, the Day One Defense Models are set to further accelerate Hawk’s current fraud offering development, enabling fraud teams to act decisively and balance response time to new threats with model precision and customer friction. With the use of Hawk’s self-service rule management, typology-specific day one models, and real-time prevention capabilities, teams will be given the opportunity to stop fraud faster. In addition, the institution’s custom anomaly detection models aim to reinforce these defenses, serving as a safety net that covers fraud teams from multiple possible threats.
