Artificial intelligence in collection solutions

What it is about

Samantha Manzo Villaseñor explained at the Third Annual Risk Conference (September 2025) how artificial intelligence (AI) can transform collections in the digital era.

The 16 slides start from the presence of AI in everyday life and in the financial sector, describe the challenges of collections in Latin America (high costs, little segmentation, low effectiveness of mass campaigns and lower contactability), explain predictive and prescriptive analytics and intelligent automation, the importance of an end-to-end digital model and the use of humanized virtual agents, and close with UNESCO recommendations for responsible use of AI.

Objectives

• Identify the challenges of traditional collections in the face of changing customer habits.
• Explain how predictive, prescriptive and conversational AI can improve recovery and customer experience.
• Set out the principles of responsible use of AI in collections.

Key takeaways

1. Collections face high traditional management costs, little segmentation and lower contactability: people answer the phone less and less.
2. Predictive analytics estimates the probability of recovery and the optimal moment for payment; prescriptive analytics assigns the best channel, tone and frequency of contact.
3. Non-payment is not always due to lack of liquidity, and customers perceiving control, for example through self-service and payment plans, improves their view of the company.
4. Humanized virtual agents can reduce hiring, training and staff maintenance costs and guide the customer through a structured flow to regularize their balance.
5. UNESCO principles are recommended: privacy and data protection, accountability, fairness and non-discrimination, transparency and explainability, and human oversight.