VynarisEarly beta Kimi K3Get your API key

Glossary · Category

Governance & compliance

131 plain-English definitions.

AI acceptable use policy
An internal policy defining what employees are and are not allowed to do with AI tools
AI accountability framework
A structure that assigns clear responsibility for the outcomes of an organization's AI systems
AI Act penalties
The fines, up to tens of millions of euros or a percentage of global revenue, imposed for violating the EU AI Act
AI audit readiness
The state of having documentation and controls in place so an organization can pass an AI compliance audit
AI audit trail
A recorded, tamper-evident log of AI system inputs, outputs, and decisions used to demonstrate compliance
AI bias audit
A formal review checking whether an AI system produces unfair or discriminatory outcomes for particular groups
AI compliance automation
Software that helps automatically track and enforce regulatory requirements across AI systems
AI compliance checklist
A step-by-step list of requirements a company must satisfy before deploying an AI system in a regulated context
AI compliance gap analysis
Comparing current AI practices against regulatory requirements to identify where a company falls short
AI compliance reporting automation
Software that automatically generates the documentation regulators require for AI system oversight
AI compliance risk score
A quantitative measure of how exposed an organization is to regulatory risk from its AI systems
AI compliance software
Tools designed to help organizations track, document, and prove compliance with AI regulations
AI compliance training for employees
Educational programs ensuring staff understand their obligations when using AI tools
AI content moderation policy
Rules governing what an AI system is permitted to generate or must filter out
AI data breach notification requirements
Legal obligations to inform regulators and affected individuals when an AI system's data is compromised
AI data governance framework
The rules and processes an organization uses to control how data is collected, stored, and used in AI systems
AI data leakage prevention
Controls that stop sensitive company or customer data from being exposed through AI tool usage
AI data privacy
Protections and controls ensuring personal or sensitive data processed by AI systems is not exposed or misused
AI ethics committee
A cross-functional group tasked with evaluating the ethical implications of proposed AI projects
AI ethics policy
An internal document setting out the principles an organization follows when building or buying AI systems
AI explainability
The degree to which a model's decisions can be understood and justified in human terms
AI export control compliance
Ensuring AI models and technology are not shared with restricted countries or entities in violation of export law
AI fairness testing
Evaluating a model's outputs across demographic groups to check for disparate treatment or impact
AI governance
The policies, processes, and oversight structures an organization uses to ensure AI systems are used safely, ethically, and in compliance with regulation
AI governance board
An internal committee responsible for reviewing and approving AI initiatives for risk and compliance
AI governance certification for employees
Formal credentials verifying that staff understand AI governance principles and requirements
AI governance consulting
External advisory services that help organizations design and implement AI governance programs
AI governance framework
A structured set of principles and controls an organization adopts to manage AI risk across its lifecycle
AI governance maturity assessment
Evaluating how developed an organization's AI governance practices are compared to best practice
AI governance policy checklist
A practical list of policy elements an organization should have in place to govern AI use
AI governance roles and responsibilities
The defined ownership structure specifying who is accountable for each aspect of AI risk and compliance
AI governance software
Platforms designed to help organizations track, document, and enforce AI policy compliance at scale
AI incident response plan
A documented procedure for detecting, containing, and remediating harm caused by an AI system failure
AI inventory management
Maintaining a complete catalog of every AI system and model in use across an organization
AI liability insurance
Insurance coverage specifically addressing financial losses caused by AI system failures or errors
AI model bias mitigation techniques
Methods used to reduce unfair or discriminatory patterns in a model's outputs
AI model bias testing tools
Software used to systematically evaluate whether a model treats different demographic groups fairly
AI model card template
A standardized fill-in-the-blank format organizations use to document their models consistently
AI model documentation template
A standardized format for recording a model's purpose, training data, and known limitations
AI model explainability regulation
Legal requirements mandating that certain automated decisions be explainable to affected individuals
AI model explainability requirements
Regulatory mandates that certain AI decisions must be explainable to the people they affect
AI model lifecycle governance
Overseeing an AI model's compliance and risk posture from initial development through retirement
AI model monitoring for compliance
Continuous tracking of a deployed model's behavior to ensure it remains within regulatory and policy bounds
AI model provenance
Documentation tracing where a model came from, how it was trained, and what data it used
AI model registry
A centralized system that catalogs every AI model an organization has deployed, along with its metadata and status
AI model risk assessment
An evaluation of the potential failure modes and harms a specific AI model could cause before it is deployed
AI model validation
Testing a model to confirm it performs as intended before it is approved for production use
AI output auditability
The ability to trace and review exactly what an AI system produced and why, after the fact
AI procurement compliance requirements
The regulatory and internal policy conditions an AI purchase must satisfy before approval
AI red teaming
Deliberately probing an AI system for vulnerabilities, harmful outputs, or security weaknesses before deployment
AI regulatory compliance tracker
A tool or system used to monitor an organization's compliance status against multiple applicable AI regulations
AI regulatory sandbox
A controlled environment where companies can test AI systems under regulatory supervision before full market release
AI risk register
A tracked list of identified AI-related risks, their likelihood, impact, and mitigation status
AI supply chain risk
The risk introduced by third-party models, datasets, or components embedded in an AI system
AI system of record
The authoritative internal record documenting which AI systems are deployed where and by whom
AI third-party audit
An independent external review verifying an AI vendor's or system's compliance and security claims
AI training data provenance
Documentation of where the data used to train a model originated and whether it was legally obtained
AI transparency requirements
Regulatory or contractual obligations that AI systems disclose how they make decisions and what data they use
AI vendor compliance certification
Third-party certifications, such as SOC 2 or ISO 42001, that an AI vendor holds to demonstrate compliance
AI vendor due diligence
The investigative process of verifying an AI vendor's claims about security, compliance, and performance before signing a contract
AI vendor GDPR compliance
Verifying that an AI vendor's data handling practices satisfy EU data protection law
AI vendor lock-in risk
The compliance and business risk created by depending too heavily on a single AI provider's proprietary systems
AI vendor risk assessment
The process of evaluating the security, compliance, and reliability risk posed by an external AI provider
AI vendor security questionnaire
A standardized set of questions procurement and security teams send AI vendors to assess their risk before purchase
AI vendor SOC 2 Type II report
An in-depth independent audit report verifying a vendor's security controls operated effectively over time
AI whistleblower channel
A confidential reporting mechanism for employees to flag AI governance or ethics violations
AI whistleblower protection
Legal protections for employees who report unsafe or noncompliant AI practices within their organization
AI whistleblowing hotline
A confidential channel for employees to report concerns about AI misuse or noncompliance
algorithmic accountability
The principle that organizations must be answerable for the decisions their algorithms make
algorithmic impact assessment
A formal evaluation of the potential societal or individual harms an AI system could cause
algorithmic transparency report
A public or regulatory disclosure explaining how an organization's automated systems make decisions
automated decision-making regulation
Laws governing when and how organizations may use algorithms to make decisions that affect individuals
California AI regulations
State-level rules in California governing the development and deployment of AI systems
CCPA AI compliance
Ensuring AI systems handling California residents' personal data comply with the California Consumer Privacy Act
Colorado AI Act
A U.S. state law regulating high-risk AI systems and requiring impact assessments
cross-border data transfer AI
Rules governing whether and how data processed by AI systems may move between countries
data localization requirements
Legal mandates that certain categories of data must be stored and processed within a specific jurisdiction
data minimization AI
The principle of collecting and processing only the data strictly necessary for an AI system to function
data processing agreement AI
A legal contract specifying how an AI vendor is permitted to process a customer's data
data protection impact assessment AI
A formal evaluation required under privacy law to assess how an AI system affects individuals' data rights
data residency AI
The requirement that data processed by an AI system stays within a specific country or region's borders
data residency compliant AI provider
An AI vendor able to guarantee that customer data stays within a specified geographic region
data sovereignty AI
The principle that data is subject to the laws of the country in which it is collected or stored, a key concern for AI deployments
data sovereignty vs data residency
The distinction between where data is physically stored and which country's laws govern it
data subject access request AI
A request from an individual to know what personal data an AI system holds and how it was used
enterprise AI compliance calendar
A schedule tracking upcoming regulatory deadlines and required compliance actions for AI systems
enterprise AI compliance dashboard
A reporting interface showing an organization's real-time compliance status across its AI systems
enterprise AI data handling policy
Rules governing how employee and customer data may be used when interacting with AI tools
enterprise AI ethics review board
An internal group that evaluates proposed AI projects for ethical concerns before approval
enterprise AI policy compliance monitoring
Ongoing automated checks that AI usage across the organization stays within approved policy
enterprise AI policy enforcement
Technical controls that automatically apply an organization's AI usage rules across all approved tools
enterprise AI policy template
A reusable starting document companies adapt to create their own internal AI usage policy
enterprise AI privacy by design
Building privacy protections into an AI system's architecture from the outset rather than adding them later
enterprise AI risk appetite
The level of AI-related risk an organization's leadership is willing to accept in pursuit of its objectives
enterprise AI risk taxonomy
A structured classification system for the different categories of risk an AI system can introduce
enterprise AI security questionnaire
A due-diligence document used to evaluate whether an AI vendor's security practices meet a company's standards
enterprise AI trust center
A vendor-published page detailing security, compliance, and privacy practices to reassure enterprise buyers
enterprise data classification AI
Categorizing company data by sensitivity level to determine which data is safe to use with which AI systems
EU AI Act compliance
The set of obligations a company must meet to legally deploy AI systems that touch the EU market
EU AI Act high-risk system
An AI application, such as one used in hiring or credit decisions, subject to the strictest obligations under EU law
GDPR AI compliance
Ensuring an AI system's handling of personal data meets the EU's General Data Protection Regulation
generative AI acceptable use policy
A specific policy defining approved and prohibited uses of generative AI tools within an organization
generative AI IP indemnification
A vendor's contractual promise to cover legal costs if their AI's output infringes someone else's intellectual property
HIPAA AI healthcare
The application of HIPAA privacy and security rules to AI tools used in healthcare settings
HIPAA compliant AI
An AI system configured and contracted so that it can legally process protected health information
human oversight requirement
A regulatory or policy mandate that humans retain meaningful control over AI decision-making
industry-specific AI regulations
Sector-specific rules, such as those in finance or healthcare, that impose additional AI compliance requirements
ISO 27001 AI systems
Applying the international information security management standard to AI infrastructure and processes
model deprecation policy
A vendor's or company's plan for retiring older AI models and migrating users to replacements
model documentation requirements
The records a company must keep describing how a model was built, trained, and validated
model drift monitoring
Ongoing tracking of whether a deployed model's performance degrades or its behavior changes over time
model explainability tools
Software that helps interpret and visualize why a model produced a particular output
model risk governance committee
An internal body responsible for overseeing and approving the risk profile of deployed AI models
model risk management
The discipline of identifying, measuring, and mitigating the risks that a model behaves incorrectly or unfairly
model risk management framework
A structured methodology banks and regulated firms use to identify and control the risks of deploying models
model risk tiering
Classifying AI models by their potential impact and risk level to apply proportionate oversight
PCI DSS AI
Ensuring AI systems that touch payment card data comply with the Payment Card Industry Data Security Standard
prompt injection risk
The security risk that malicious input tricks an AI system into ignoring its instructions or leaking data
right to explanation
A legal or regulatory principle giving individuals the right to understand how an automated decision about them was made
shadow AI detection
Tools and techniques used to discover unauthorized AI tool usage across an organization's network
shadow AI risk
The security, compliance, and data leakage exposure created when employees use unapproved AI tools
SOC 2 compliance AI vendor
Verifying that an AI vendor has passed an independent audit of its security controls
SOC 2 for AI
Extending the SOC 2 security and compliance audit framework to cover AI systems and data handling
sovereign AI
AI infrastructure and models built, hosted, and operated entirely within a nation's own borders and legal control
Standard Contractual Clauses AI
Legal templates used to lawfully transfer personal data outside the EU when using AI vendors
third-party AI risk management
Managing the risks introduced when a company relies on external AI vendors or models
whistleblower AI misuse reporting
Channels and protections for employees reporting concerns about improper AI system use