cyber.hendryadrian.com
Security News · darkreading.com

F-Secure Maps 26 AI-Related Consumer Harms and Where to Defend Against Them

· Original article ↗

Summary

F-Secure’s AI Harms Taxonomy maps 26 consumer harms across five relationship types and 19 intermediary layers, with 24 defenses. It highlights hidden malicious instructions and the shared responsibility of device, AI and service providers.

Key points

  • The taxonomy maps 26 consumer harms across five relationship types and 19 intermediary layers, and identifies 24 defenses.
  • Examples include AI-generated voice or video impersonation, harmful financial decisions by agents, personal-information exposure and actions based on outdated context.
  • Malicious instructions hidden in content processed by an AI may influence an agent’s actions or a chatbot’s responses.
  • The research says responsibility for safeguards is shared among users’ device security, AI providers and companies operating connected services.
  • F-Secure says AI agents with access to accounts and linked services can cause consequences without the user being part of each interaction.

Article Details

Event Type
F-Secure announced an AI Harms Taxonomy research framework mapping consumer harms associated with AI systems.
Impact
The taxonomy maps 26 consumer harms across five relationship types and 19 intermediary layers, with 24 defenses. It identifies risks including AI-generated voice and video impersonation, harmful financial decisions by agents, personal-information exposure, outdated context, and malicious instructions hidden in content processed by AI. The research emphasizes that responsibility for safeguards is shared among device-security providers and the companies building and operating AI systems.

People

Dr. Megan SquirePrincipal Threat Intelligence Researcher at F-Secure who developed the AI Harms Taxonomy.

Vendors

F-Securecommunicate, decide and run our digital lives. But when it goes wrong, how does it actually hurt people? F-Secure’s new AI Harms Taxonomy maps 26 distinct harms to consumers, tracing the path each one takes to reach a

Products

TrustGuardQuTwo, is the company’s work on protecting people in a world where AI agents increasingly act for them. TrustGuard is designed to help protect people during AI interactions, while TrustPath aims to secure their digitalTrustPathincreasingly act for them. TrustGuard is designed to help protect people during AI interactions, while TrustPath aims to secure their digital journeys from discovery to delivery. The AI Harms Taxonomy gives that work

Related Articles