다음을 통해 공유


AttackStrategy Struct

Definition

Strategies for attacks.

public readonly struct AttackStrategy : IEquatable<Azure.AI.Projects.AttackStrategy>
type AttackStrategy = struct
Public Structure AttackStrategy
Implements IEquatable(Of AttackStrategy)
Inheritance
AttackStrategy
Implements

Constructors

AttackStrategy(String)

Initializes a new instance of AttackStrategy.

Properties

AnsiiAttack

Utilizes ANSI escape sequences to manipulate text appearance and behavior.

AsciiArt

Generates visual art using ASCII characters, often used for creative or obfuscation purposes.

AsciiSmuggler

Conceals data within ASCII characters, making it harder to detect.

Atbash

Implements the Atbash cipher, a simple substitution cipher where each letter is mapped to its reverse.

Base64

Encodes binary data into a text format using Base64, commonly used for data transmission.

Baseline

Represents the baseline direct adversarial probing, which is used by attack strategies as the attack objective.

Binary

Converts text into binary code, representing data in a series of 0s and 1s.

Caesar

Applies the Caesar cipher, a substitution cipher that shifts characters by a fixed number of positions.

CharacterSpace

Alters text by adding spaces between characters, often used for obfuscation.

CharacterSwap

Swaps characters within text to create variations or obfuscate the original content.

Diacritic

Adds diacritical marks to characters, changing their appearance and sometimes their meaning.

Difficult

Represents a default set of difficult complexity attacks. Difficult complexity attacks include attacks that require access to significant resources and effort to execute an attack such as knowledge of search-based algorithms in addition to a generative AI model.

Easy

Represents a default set of easy complexity attacks. Easy complexity attacks require less effort, such as translation of a prompt into some encoding, and does not require any Large Language Model to convert or orchestrate.

Flip

Flips characters from front to back, creating a mirrored effect.

Jailbreak

Injects specially crafted prompts to bypass AI safeguards, known as User Injected Prompt Attacks (UPIA).

Leetspeak

Transforms text into Leetspeak, a form of encoding that replaces letters with similar-looking numbers or symbols.

Moderate

Represents a default set of moderate complexity attacks. Moderate complexity attacks require having access to resources such as another generative AI model.

Morse

Encodes text into Morse code, using dots and dashes to represent characters.

ROT13

Applies the ROT13 cipher, a simple substitution cipher that shifts characters by 13 positions.

StringJoin

Joins multiple strings together, often used for concatenation or obfuscation.

SuffixAppend

Appends an adversarial suffix to the prompt.

UnicodeConfusable

Uses Unicode characters that look similar to standard characters, creating visual confusion.

UnicodeSubstitution

Substitutes standard characters with Unicode equivalents, often for obfuscation.

Url

Encodes text into URL format.

Methods

Equals(AttackStrategy)

Indicates whether the current object is equal to another object of the same type.

ToString()

Returns the fully qualified type name of this instance.

Operators

Equality(AttackStrategy, AttackStrategy)

Determines if two AttackStrategy values are the same.

Implicit(String to AttackStrategy)

Converts a string to a AttackStrategy.

Implicit(String to Nullable<AttackStrategy>)
Inequality(AttackStrategy, AttackStrategy)

Determines if two AttackStrategy values are not the same.

Applies to