Looking at the development of AI technology and increasing pervasiveness in the financial and digital space, it is critical for financial institutions to consider how to make machine learning as ethical as possible. Implementing machine learning elements into fraud scoring systems can enhance customer experience and overcome challenges faced by banks, but it is essential that the technology to avoids inherent bias, discrimination and model degradation for continued accuracy.
To implement ethical AI in fraud scoring, companies are looking to build governance models in machine learning that can account for built-in biases and remove them. Self-learning machine learning models for banks using new information also works to future-proof the institutions. Improved machine learning technology is perfectly suited to the needs of tier two and tier three banks which require an edge where they lack data science experts.
Making ethical AI accessible to smaller banks through a network of data available when identifying risks in transactions allows the development of new models and products with more accuracy and less bias for fraud scoring.
Sign up for this Finextra webinar, hosted in association with ACI, to join our panel of industry experts as they discuss the following areas:
- How can ethical AI be achieved in fraud scoring?
- What built-in biases are present in current machine learning models and how can they be avoided or overcome?
- How do we overcome machine learning model degradation?
- How will tier two and tier three banks benefit from ethical AI models?
- How can ethical AI create a more consistent customer experience for users?
- How should financial institutions implement ethical AI technologies to future-proof their operations?
- Gary Wright - Head of Research, Finextra [Moderator]
- Patricia Rojas - Senior Manager - Data Scientist, ACI Worldwide