The author further stresses the need for accountability when mistakes occur due to AI, suggesting that businesses should have someone on the leadership team to provide oversight. The article concludes by stating that while laws and regulations will eventually dictate certain aspects of AI, businesses must be ready to absorb the risks and consequences of deploying AI and remain vigilant in monitoring and improving their AI models.
Key takeaways:
- Protecting individuals and their data is at the core of responsible AI use, especially when it comes to financial information. Businesses should work with data experts who can keep customers' private information safe while using payment data to enable personalized service.
- AI can yield biased outcomes by relying on data or algorithms that introduce unintended bias. To prevent this, businesses should choose the most accurate, audited and unbiased data sources available and use a system of audits during the development and implementation of the machine learning model.
- While AI can automate many functions related to bill payment, not everything related to bill payment is right for full-scale automation. A balance between automation and human augmentation in customer service is necessary, especially when it comes to important tasks related to a consumer’s finances.
- Companies that add AI to their payment processes need to maintain a “buck stops here” attitude if and when errors are made. They should appoint someone on the leadership team to provide accountability and oversight, or talk with their payments technology provider about providing that service.