Hidden Capabilities

Hidden capabilities refers to behaviors or abilities that an AI system possesses but that were not disclosed in its documentation, testing results, or intended use, and that may only surface under particular conditions or prompts. For governance teams, this matters because a system's risk profile is only as complete as its disclosed capabilities, so undocumented functions can create liability, safety gaps, or regulatory exposure that standard reviews miss. Tracking news on this topic helps compliance and risk leaders judge whether their vendor assessments, pre-deployment testing, and ongoing monitoring are looking for the right things.

1 item

Research2026-10-09

Research Proves AI Models Can Hide Capabilities That No Inspection Can Find

Researchers from the University of Oxford have published a paper on arXiv showing that AI models can contain hidden capabilities activated only by a secret trigger. These hidden functions cannot be detected through weight inspection, fine-tuning, or any other currently available technique. The findings challenge the assumption that examining a model's internal structure is sufficient to verify its safety. Organizations that rely on pre-deployment model evaluation as a safety guarantee must now treat that guarantee as incomplete.

model evaluation hidden capabilities backdoor AI vendor due diligence open-weight models EU AI Act University of Oxford