MIT AI Risk Repository · domain 7: AI system safety, failures, & limitations

7.4 Lack of transparency or interpretability

Challenges in understanding or explaining the decision-making processes of AI systems, which can lead to mistrust, difficulty in enforcing compliance standards or holding relevant actors accountable for harms, and the inability to identify and correct errors.

Risk entries
42
Frameworks citing it
12
Recorded incidents
5
Incidents since 2020
2
Causal entity (risk entries)
Causal entity (risk entries) 21 0 AI: 21 AI 21 Other: 12 Other 12 Human: 8 Human 8 Not coded: 1 Not coded 1
Causal entity (risk entries)
LabelValue
AI21
Other12
Human8
Not coded1
Intent (risk entries)
Intent (risk entries) 23 0 Unintentional: 23 Unintentional 23 Other: 18 Other 18 Not coded: 1 Not coded 1
Intent (risk entries)
LabelValue
Unintentional23
Other18
Not coded1
Timing (risk entries)
Timing (risk entries) 21 0 Post-deployment: 21 Post-deployment 21 Other: 14 Other 14 Pre-deployment: 6 Pre-deployment 6 Not coded: 1 Not coded 1
Timing (risk entries)
LabelValue
Post-deployment21
Other14
Pre-deployment6
Not coded1
Recorded incidents per yearIncident date; current year partial
Recorded incidents per year 1 0 2016: 1 2016 1 2017: 1 2017 1 2018: 1 2018 1 2021: 1 2021 1 2022: 1 2022 1
Recorded incidents per year
LabelValue
20161
20171
20181
20211
20221
Entries by levelRisk categories, subcategories and additional evidence coded to this subdomain
Entries by level 23 0 Risk Category: 23 Risk Category 23 Risk Sub-Category: 19 Risk Sub-Category 19
Entries by level
LabelValue
Risk Category23
Risk Sub-Category19
  • Lack of transparency, explainability, and trust

    "Understanding how AI reaches conclusions or why AI systems perform specific actions motivates an entire branch of interpretability research [111], but physical embodiment raises the stakes for unders...

    Embodied AI: Emerging Risks and Opportunities for Policy Action (Perlo2025) · Other · Unintentional · Other

  • Accountability

    An essential feature of decision-making in humans, AI, and also HLI-based agents is accountability. Implementing this feature in machines is a difficult task because many challenges should be consider...

    A Survey of Artificial Intelligence Challenges: Analyzing the Definitions, Relationships, and Evolutions (Saghiri2022) · AI · Unintentional · Other

  • Transparency

    an external entity of an AI-based ecosystem may want to know which parts of data affect the final decision in a learning model

    A Survey of Artificial Intelligence Challenges: Analyzing the Definitions, Relationships, and Evolutions (Saghiri2022) · AI · Other · Post-deployment

  • Reproducibility

    How a learning model can be reproduced when it is obtained based on various sets of data and a large space of parameters. This problem becomes more challenging in data-driven learning procedures witho...

    A Survey of Artificial Intelligence Challenges: Analyzing the Definitions, Relationships, and Evolutions (Saghiri2022) · AI · Unintentional · Post-deployment

  • Verifiability

    In many applications of AI-based systems such as medical healthcare and military services, the lack of verification of code may not be tolerable... due to some characteristics such as the non-linear a...

    A Survey of Artificial Intelligence Challenges: Analyzing the Definitions, Relationships, and Evolutions (Saghiri2022) · AI · Unintentional · Post-deployment

  • Insufficient AI development documentation

    "Throughout the development of an AI system, it is vital to document every decision and action taken. This is not only essential to optimize the development process itself but also required for the au...

    AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024) · Human · Other · Pre-deployment

  • Inappropriate degree of transparency to end users

    "The transparency to end users of the AI system increases the user’s trust in the AI application. If not adequately integrated into the design, this might prevent the proper operation and cause potent...

    AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024) · Human · Other · Pre-deployment

  • Lack of explainability

    "The explainability of AI systems based on so-called black-box models is often limited. This opaqueness of AI systems can prevent developers from detecting shortcomings in the data or the model itself...

    AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024) · Other · Other · Other

  • Concept drift

    "Concept drift refers to a change in the rela- tionship between input variables and model output. If not treated appropriately, concept drift can reduce the reliability of AI systems."

    AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024) · Other · Other · Other

  • Explainability & Transparency

    "The feasibility of understanding and interpreting an AI system's decisions and actions, and the openness of the developer about the data used, algorithms employed, and decisions made. Lack of these e...

    AI Risk Profiles: A Standards Proposal for Pre-Deployment AI Risk Disclosures (Sherman2023) · AI · Other · Other

  • Degree of Transparency and Explainability

    "Transparency is the characteristic of a system that describes the degree to which appropriate information about the system is communicated to relevant stakeholders, whereas explainability describes t...

    Sources of Risk of AI Systems (Steimers2022) · AI · Unintentional · Post-deployment

  • Risks from models and algorithms (Risks of explainability)

    "AI algorithms, represented by deep learning, have complex internal workings. Their black-box or grey-box inference process results in unpredictable and untraceable outputs, making it challenging to q...

    AI Safety Governance Framework (TC2602024) · AI · Unintentional · Other

  • Accountability

    "The ability to determine whether a decision was made in accordance with procedural and substantive standards and to hold someone responsible if those standards are not met."

    An Exploratory Diagnosis of Artificial Intelligence Risks for a Responsible Governance (Teixeira2022) · Other · Other · Other

  • Opacity

    "Stems from the mismatch between mathematical optimization in high-dimensionality characteristic of machine learning and the demands of human-scale reasoning and styles of semantic interpretation."

    An Exploratory Diagnosis of Artificial Intelligence Risks for a Responsible Governance (Teixeira2022) · AI · Unintentional · Post-deployment

  • Explainability

    "Any action or procedure performed by a model with the intention of clarifying or detailing its internal functions."

    An Exploratory Diagnosis of Artificial Intelligence Risks for a Responsible Governance (Teixeira2022) · AI · Other · Post-deployment

  • Complexity-induced knowledge gap

    "The complexity of AI models and systems makes it challenging to demonstrate harm or establish a clear causal link between AI actions and their consequences."

    A Taxonomy of Systemic Risks from General-Purpose AI (Uuk2025) · Other · Unintentional · Other

  • Opaque AI networks

    "The complexity and opacity of AI models and systems make it difficult to predict and manage their behavior."

    A Taxonomy of Systemic Risks from General-Purpose AI (Uuk2025) · Other · Other · Post-deployment