SECAI Core · phase 15 of 15 · noun

Model skewing

When the behavior of a machine learning model is altered through biased or manipulated training or feedback data

The Explain card

Plain English
Model skewing is when a model's behaviour is altered by biased or manipulated training or feedback data, pushing its decisions in a direction the attacker wants.
Example
A spam filter retrains on user feedback. An attacker creates many accounts that mark their spam as "not spam", and over weeks the filter learns to let their campaigns through.
Why it matters
Any model that learns from production data or user feedback can be steered by whoever controls that data. Defenders must validate feedback, limit its influence and watch for behavioural shifts.
Hook
Feed the model lies long enough and it believes them.

Word knowledge

How the word is built, where it came from, and what it sits beside in memory.

In a sentence

The security team detected model skewing after a compromised sensor gateway supplied altered temperature readings to the nightly retraining pipeline.

Why these words
  • model French modèle, through Italian modello, from Latin modulus, small measure identifies the system affected
  • skewing English skew plus -ing, from Middle English skewen and Old North French eskiuer, to turn aside acts as the head noun naming the process
Where it came from
Origin
An English technical compound formed from model and skewing
Entered the language
21st century
What changed
Model broadened from a physical representation to a computational system, while skew developed from turning obliquely to causing distortion; skewing then became an action noun.
How it is spelled
Pattern
The phrase is an open compound, written as two words without a hyphen.
Pattern
Skewing adds -ing directly to skew and retains the final w.
Breaks the pattern
In skew, ew represents the sound /juː/ in many English accents.

Spelled like

  • model training
  • model monitoring
  • chewing
Broken into chunks
  • model
    • module
    • modulus
  • skew
    • askew
    • eschew
  • ing
    • training
    • poisoning
What it sits beside

Same subject

  • data poisoning
  • label flipping
  • adversarial machine learning
  • training-data integrity

Same shape

model-targeting attacks

training-data attacks

Where it sits in the deck

Phase 15: MLOps, Continuous Delivery, and Operational Resilience

Close the loop — operational practices, drift management, and delivery discipline that keep deployed AI systems healthy, current, and continuously improving.

Neighbours in this phase