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

Palantir

Palantir deploys AI as an enterprise and government platform (AIP/Foundry) that integrates and operationalises third-party LLMs from OpenAI, Anthropic, Meta, Google, and xAI in secure environments, including classified government cloud systems. It has not built a proprietary foundational model; its role centres on deployment, operationalisation, and controls.

We reviewed 3 public documents from Palantir and found it acknowledges 10 of 24 AI risks, including capability & robustness and discrimination.

Sector: Information

Roles: AI deployer

AI risks Palantir acknowledges

  • Capability & robustness: “AI systems are already in wide, and rapidly growing, use... and they can and do fail in many different ways... leading to real harms... provisions to address the inherent brittleness of AI technologies... the tendency of models be anti-resilient to changes in data, environment, application context, etc.” (Palantir Australia Response to Supporting Responsible AI ...)
  • Discrimination: “applying facial recognition technology to predict sexual orientation from facial features ... implicates sensitive category inferences that are morally, scientifically, and legally questionable ... potentially negative feedback loops, and subtly compounding disparate impacts” (Palantir AI Ethics)
  • Loss of privacy: “The notice-and-consent paradigm for managing privacy risks has failed in the face of complex and sophisticated uses and transfers of personal information...” (Palantir Australia Response to Supporting Responsible AI ...)
  • Overreliance & unsafe use: “They should also take account of the propensity of such models to induce levels of end-user trust that exceed the system's competence.” (Palantir Australia Response to Supporting Responsible AI ...)
  • Fraud & scams: “AI, including our use of AI, may create additional cybersecurity risks... and may result in... phishing attacks... As AI capabilities improve... we may see cyberattacks crafted with an AI tool... by creating more effective phishing emails or social engineering and by exploiting vulnerabilities... utilizing false image or voice recognition.” (2024 10-K Palantir)
  • Loss of human agency: “AI is both most effective and most defensible when employed to assist and enhance human execution and decision making rather than replacing it... the limits of AI must be acknowledged... to ensure moral agency and culpability.” (Palantir AI Ethics)
  • Disinformation & influence: “Such systems have swiftly been adopted into workflows - both productive and malicious (e.g., producing misinformation at previously unprecedented quality and scale).” (Palantir Australia Response to Supporting Responsible AI ...)
  • Weapons & cyberattacks: “In addition, enhanced risk of cyberattacks could be a result of us or a third party incorporating the output of an AI tool, such as malicious code from an AI generated source code.” (2024 10-K Palantir)
  • Governance failure: “accountability mechanisms are likely to be insufficient or even counter-productive unless there is clarity about who is ultimately responsible for the actions of an AI system... it can be unclear who is ultimately accountable for the negative implications or potential harms of AI system use” (Palantir Australia Response to Supporting Responsible AI ...)
  • False information: “Third-party AI capabilities that can be integrated with our platforms could also produce false or "hallucinatory" inferences” (2024 10-K Palantir)

Documents we reviewed