Adversarial attack

AIGP glossary · Last reviewed: 2026-07-08 · By Victor Humenhuk (AIGP certified)

Adversarial attack - Deliberately crafted inputs designed to fool a model into errors or unsafe behaviour.

In the AIGP body of knowledge, Adversarial attack comes up under Module 3: Governance & Risk Management · BoK III.A; Module 8: AI Governance Vocabulary.

Adversarial attack in context

• The [[Security risk]] card carries the most testable vocabulary → [[Adversarial attack]]s, [[Hallucination]]s, [[Deepfake]]s and [[Data poisoning]]. (The four AI risk categories)

Where Adversarial attack is covered in the AIGP study notes

• The four AI risk categories
• Risks, security and harms

Related terms

• Operational risk
• Legal risk
• Security risk
• Privacy risk
• Data poisoning
• Deepfake
• Hallucination
• Data persistence
• Data spillover
• Data leak

Test yourself on Adversarial attack

Recognising a definition is not the same as applying it in an exam scenario. Work through the free AIGP practice questions, or read the full AIGP study guide - every study note is free.

← Back to the AIGP glossary