Which of the following BEST describes the risk of model staleness in a deployed machine learning system?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — model staleness means your deployed model gets less accurate over time because real-world conditions drift away from what it learned during training. You're not dealing with malware, data leakage, or compliance violations here.
Full Explanation
The option saying predictive accuracy degrades as real-world conditions diverge from training conditions is correct because model staleness, or drift, is the natural loss of accuracy over time as customer behavior, market dynamics, and language use move beyond the training data. The malicious-code option describes a supply chain or insider threat. The sensitive-data-leak option describes model inversion or membership inference risk. The regulatory-threshold option describes a compliance failure that may or may not relate to staleness.