Company AI Risk Explorer
Loading the data…
Welcome to the Company AI Risk Explorer

This tool maps what large global companies say in public about AI risk. We reviewed public documents from roughly 200 of the world's largest companies. We coded every risk they acknowledge against the AI Risk Repository taxonomy.

Filter by company, sector, AI role, or risk. Compare what companies acknowledge against expert ratings of severity, sector vulnerability, and role responsibility.

← Back to the overview Company AI Risk Explorer

Companies × risks

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.

25 of the world's largest companies we reviewed acknowledge the AI risk "Transparency & interpretability", including IBM, SAP and Mastercard.

Which companies acknowledge this risk?

  • IBM: “Explanations for model output decisions might be difficult, imprecise, or not possible to obtain... Without clear explanations for model output, it is difficult for users... to understand and trust the model.” (Foundation models: Opportunities, risks and mitigations)
  • SAP: “We know less about how decisions are ultimately made when it comes to AI than we do when it comes to conventional software," Wieczorek says.” (The Ethical Side of AI at SAP | SAP News Center)
  • Mastercard: “Inscrutable models do not allow you to "walk back" the logic they used to make a decision, which can become a major issue where critical decisions result in mistakes, regulatory non-compliance, or financial loss.” (Mastercard Generative AI Cybersecurity Report)
  • Alphabet (Google): “being able to tell whether an AI's answers are good or bad... can become challenging when the AI has advanced capabilities.” (Taking a responsible path to AGI)
  • American Express: “can make it difficult to assess proper operation, reduce error, or understand and explain their outputs” (American Express 2024 Annual Report)
  • Baidu: “The Baidu Technology Ethics Committee aims to reduce the "black box" risks of AI algorithms and enable more anticipatory AI governance.” (Baidu 2025 Environmental, Social and Governance Report)
  • Citigroup: “most of the companies that develop the models do not disclose which data was used to train models, making it impossible for users to verify the training data... this lack of transparency forces users of generative AI to blindly trust their claims.” (Delivering Improved Experience Through Generative AI)
  • Coinbase: “One of the key challenges of AI is that its decisions are not transparent” (Blockchain and Artificial Intelligence (AI): Complementary Technologies That Can Make Each Other Better)
  • Eli Lilly: “The algorithms and models utilized in AI systems may have limitations, including biases, errors, or inability to handle certain data types or scenarios or to render explainable outputs.” (Eli Lilly 2024 Annual Report)
  • Johnson & Johnson: “When it is not possible to fully explain how an AI model works, we seek to design and monitor the AI model proactively...” (Doing the Right Thing - AI & Ethics - Johnson & Johnson)
  • Novartis: “ensuring transparency... addresses key risk areas such as transparency, explainability” (Our commitment to ethical and responsible use of Artificial ...)
  • T. Rowe Price: “Another risk is around transparency. AI models can be complex and opaque, making it difficult to trace the basis of a response.” (Using Artificial Intelligence to Enhance Our Investment Processes | T Rowe Price)
  • Alibaba: “The black box of deep learning also creates challenges for system optimization and human intervention.” (Alibaba's CTO On Everything You Wanted To Know About AI Ethics (But Were Too Afraid To Ask))
  • Alnylam Pharmaceuticals: “mitigate the risks and challenges associated with AI, such as ... transparency, explainability” (Responsible Use of Artificial Intelligence)
  • AstraZeneca: “generate insights which are difficult to interpret and explain” (AstraZeneca Risk Supplement 2024)
  • Cisco: “Often, it is not clear to users when and how AI is used or involved in decision- making.” (Cisco Principles for Responsible Artificial Intelligence)
  • Goldman Sachs: “the complexity of many AI models makes it challenging to understand why they are generating particular outputs. This limited transparency increases the challenges associated with assessing the proper operation of AI models” (Goldman Sachs 2024 Form 10-K)
  • Merck: “the inner workings of AI can be difficult to understand” (Merck's commitment to developing and using artificial intelligence ...)
  • Oracle: “AI technologies are rapidly changing and present evolving legal, regulatory and ethical issues... including... a perceived lack of transparency... any of which could expose us or our customers to reputational or legal risk and inhibit adoption of our AI products.” (Oracle Form 10-K 2025 Annual Report)
  • Philip Morris International: “This workstream will aim to mitigate, amongst others, risks such as: ... Lack of transparency and explainability” (The state of AI at PMI | PMI - Philip Morris International)
  • PNC Financial Services: “because of the complex way in which the many variables in AI/machine learning models interact, the results of these models are often less interpretable than traditional statistical models.” (PNC Financial Services 2025 Annual Report)
  • Prosus: “Such inherent risks range from a lack of explainability of certain decisions” (Prosus Annual Report 2025)
  • Walmart: “aimed at mitigating risks such as unfair outcomes, lack of transparency” (FY2025 ESG Report | Walmart Corporate)
  • Wells Fargo: “the complexity and limited transparency of certain AI models can make it challenging to understand why they are generating particular outputs.” (Wells Fargo 2025 Annual Report)
  • Wolters Kluwer: “AI can often feel like a "black box.” (Responsible and reliable AI in legal: Wolters Kluwer's principles)