Decoder AI & Data Terms

Data Poisoning

STANDARD DEFINITION · CACHED

is a type of in which a malicious actor injects manipulated data into, or modifies existing data within, a to compromise a model’s performance or integrity. Objectives include indiscriminate performance degradation through , , and the creation of a that causes attacker-chosen behavior when specific triggers appear during . Other integrity failures can include systematically biased or otherwise manipulated outputs. Poisoning influences what a model learns during or , potentially undermining the of models, including systems and . This threat is particularly concerning in systems and environments when training data comes from untrusted contributors.

[PREVIEW MODE] Definitions streamed at five depths, references, and related terms are available to signed-in readers — [SIGN IN]

Contextual terminology map

KEEP IN VIEW

Concept inspired by Dev Valladares’s Infinite Wiki. Independently built; not affiliated with or endorsed by the original.

Keyboard shortcuts

On. Switch them off if they collide with your assistive tools; ? still opens this sheet.

Go to

Press g then the letter.

  • gh Latest
  • gs Sources
  • gm Media
  • gv Videos
  • gp Podcasts
  • gc Calendar
  • gd Decoder
  • gz Dataviz
  • ga Datasets
  • gb Blog
  • gk Markets
  • gj Careers
  • gn Prompt Notebook

On this page

  • / Focus search, where there is one
  • t Back to top
  • ? This list
  • Esc Close