Contract intelligence for commercial lending

Source JPMorgan Chase (COiN) — public case. This is an industry example, not our project
360,000
hours of annual manual review eliminated
~12,000
credit agreements processed in seconds

The problem

Interpreting commercial-loan agreements consumed enormous, expensive lawyer and loan-officer time and was prone to human error.

The AI approach

COiN (“Contract Intelligence”) uses machine learning and natural-language processing (with image recognition) to extract ~150 standardised attributes from credit agreements in seconds, on the bank’s private cloud, feeding outputs into existing workflows.

Evidence it works

Bloomberg reported COiN eliminated ~360,000 hours of annual manual review, processing ~12,000 commercial credit agreements in seconds, and reduced loan-servicing errors attributable to human mistakes.

What “good” looks like

Faster review, consistent rule application, fewer servicing errors, and lawyers redeployed to higher-value judgement work.

Feasibility & cost shape

Significant up-front modelling and integration; pays off where document volume is high and contract structure is repeatable.

Our independent view

The canonical document-intelligence case. The transferable principle is to automate the standardised 80% and route the ambiguous tail to people — not to aim for full automation.

Source & attribution

Based on publicly reported information about the JPMorgan Chase (COiN) work.

This is an industry example included for illustration. It is not a Leia Intelligence project, and no client of ours is implied. Figures are as publicly reported by the original parties.

Sources: Bloomberg ("JPMorgan Software Does in Seconds What Took Lawyers 360,000 Hours") · The Independent · JPMorgan 2016 Annual Report

This is the class of problem we help organisations tackle — book a call.
Book a call