MENTAL MODEL · M0841

Black Box Algorithms

DecideHigh supportDecision Science

Included

Version 1.0.0 · Updated 2026-07-30

CORE DEFINITION

The decision-making process of deep learning models is so complex that even developers cannot explain why the model reaches a certain conclusion (e.g., why a loan is denied, why a sentence is given).

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

Demand explainability. When introducing AI decisions, without attribution, there is no accountability. As a scaffold, it reminds us to use end-to-end models cautiously in high-risk domains, or to pair them with explainability tools (XAI).

anchor

Anchor fast decisions

Refers to the opaque internal decision-making process of complex models (especially deep learning), making it difficult to explain how inputs lead to outputs, resulting in accountability challenges.

MINIMUM ACTION

In progress 0/3

Practice this model in one real situation:

Distinguish between interpretable models and black boxes
use interpretable methods or post-hoc explanations for high-risk scenarios
establish auditing and traceability.

Check to track your progress (stored locally)

Learning progress0%

account_treeGenealogyexpand_more

menu_bookReferencesexpand_more

Source support: Explicit

link
baike.baidu.comhttps://baike.baidu.com/item/%E7%AE%97%E6%B3%95%E9%BB%91%E7%AE%B1/50902717ZH · Explicit
verified

RELATED MODELS