Welcome to the Enkrypt AI Red Teaming documentation
bias_test
- Identifying and exposing biased outputs.
cbrn_test
- Addressing vulnerabilities related to chemical, biological, radiological, and nuclear domains.
harmful_test
- Eliciting responses that promote harm or danger.
insecure_code_test
- Producing insecure or harmful code snippets.
toxicity_test
- Generating harmful or offensive content.
pii_test
- Exposing personally identifiable information.
copyright_test
- Exposing copyrighted material.
misinformation_test
- Exposing misinformation.
system_prompt_extractions_test
- Exposing system prompt extractions.
sponge_test
- Exposing infinite loops.
competitor_test
- Exposing information about competitors.
custom_test
- Custom dataset test.
adv_bias_test
- Uncovering biased outputs through adversarial methods.
adv_info_test
- Extracting sensitive or unintended information from a generated dataset.
adv_tool_test
- Misusing integrated tools or features.
adv_command_test
- Manipulating the model to execute unintended commands.
adv_pii_test
- Exposing personally identifiable information.
adv_competitor_test
- Extracting confidential information about competitors.
alignment_and_governance_test
- Testing the alignment and governance of the model.
input_and_content_integrity_test
- Testing the integrity of the input and content.
infrastructure_and_integration_test
- Testing the infrastructure and integration of the model.
security_and_privacy_test
- Testing the security and privacy of the model.
human_factors_and_societal_impact_test
- Testing the human factors and societal impact of the model.
access_control_test
- Testing the access control and permissions of the model.
physical_and_actuation_safety_test
- Testing the physical and actuation safety of the model.
reliability_and_monitoring_test
- Testing the reliability and monitoring of the model.
governance_test
- Testing the governance of the model.
agent_output_quality_test
- Testing the quality of the agent’s output.
tool_misuse_test
- Testing the misuse of the model’s tools.
privacy_test
- Testing the privacy of the model.
reliability_and_observability_test
- Testing the reliability and observability of the model.
agent_behaviour_test
- Testing the behaviour of the agent.
access_control_and_permissions_test
- Testing the access control and permissions of the model.
tool_extraction_test
- Test if the agent tool infromation can be extracted in outputs.