Arum: Using AI to transform collections system delivery

Key Takeaways#

• AI is transforming collections and recovery systems by improving speed, accuracy, and cost-effectiveness.
• AI enables more efficient requirements gathering by analysing legacy documents and producing initial business requirement drafts.
• Compliance risks can be reduced through AI tools that securely handle sensitive and confidential client data.
• AI can benchmark requirements against established best practices to identify gaps early.
• During design, AI supports validation of configuration and logic, reducing inconsistency and error risk.
• AI improves documentation quality, aiding maintainability and lowering future support costs.
• Testing timelines can be significantly shortened through AI-generated test scripts and enhanced validation.
• AI supports technical testing activities, including data reconciliation and migration accuracy.
• Strong AI governance is essential, including private models and expert review of outputs.
• The “mask, share, generate, restore” framework enables responsible AI use without compromising data security.
• Future AI use cases include change control support, automated training content, and post-implementation monitoring.
• System change programmes should start with governance and focus on friction points where AI adds value.

Key Statistics#

• A library of over 500 best-practice standards has been developed based on years of delivery experience.

Key Discussion Points#

• The role of AI in improving customer engagement and delivery of change programmes.
• The importance of establishing a strong AI governance framework from the outset.
• How AI accelerates and simplifies the requirements gathering process.
• Practical outcomes from recent engagements demonstrating AI-driven testing benefits.
• AI’s ability to identify design inconsistencies that may create future issues.
• The use of AI-generated documentation to support onboarding and handover.
• Maintaining regulatory compliance while deploying AI tools.
• AI-generated test scripts and their value in User Acceptance Testing.
• The potential for AI to support training and outcome monitoring after go-live.
• Emerging applications of AI in change control processes.
• How governance structures and data masking build trust in AI usage.
• Focusing AI adoption on practical applications that reduce friction and unlock value.

Podcast Description#

This episode of Arum AI Insights examines the impact of artificial intelligence and large language models on collections and recovery. Hosted by Ella and featuring Owen Atkinson, Head of Delivery at Aram Global, the discussion explores how AI is reshaping system implementations, upgrades, and migrations. The episode focuses on efficiency, accuracy, compliance, and governance, using real-world examples to illustrate how organisations can responsibly embed AI within delivery teams to achieve stronger outcomes.