MLARM case studies
MLARM · Put your knowledge to work in context.
The Machine Learning and Artificial Intelligence Risk Management (MLARM) credential is designed for professionals who need to identify, assess, and mitigate risks introduced by AI and ML systems. It is relevant for risk managers, model validators, compliance officers, data scientists, and auditors working in regulated industries such as finance, healthcare, and technology. The MLARM case studies are 5 cases and about 2,520 words.
1The Champion-Challenger Calibration Drift
A mid-sized bank's probability-of-default model for SME lending has been in production for 14 months. Development validation reported AUC 0.81 and a predicted one-year default rate of 1.2%.
574 words, about 3 min2Vendor Scorecard and Missing Adverse Action Reasons
LendBridge, a fintech lender, uses a vendor score called ScoreNu for small-business credit decisions. ScoreNu is a proprietary gradient-boosting model; the contract provides only a score and four generic reason codes, such as 'credit history.' LendBridge combines ScoreNu with ten internal features, but the vendor score carries roughly 65% of decision weight.
514 words, about 3 min3Synthetic Fraud Data with a Privacy Leak Signal
PayFleet, a payments processor, sees fraud in 0.04% of 2.1 million monthly transactions, about 840 fraud cases. To improve a rare-event fraud model, the data science team trains a gradient-boosted model only on synthetic data generated by CTGAN: 50,000 synthetic fraud and 200,000 synthetic non-fraud records.
477 words, about 2 min4Fairness Metric Conflict in Hiring Screening
A technology company uses a machine-learning model to screen 38,000 applicants for engineering roles and invites the top 15% to interview. The applicant pool is 22% female and 9% underrepresented minority.
482 words, about 2 min5GenAI Assistant Expansion and Hallucination Risk
A retail bank deploys a vendor-hosted large language model assistant for 1,200 contact-center agents. The assistant answers questions about fees, disputes, and account closures.
473 words, about 2 min