0 / 0
Confidential information in data risk for AI

Confidential information in data risk for AI

Risks associated with input
Training and tuning phase
Intellectual property
Traditional AI risk


Models might be trained or fine-tuned using confidential data or the company’s intellectual property, which could result in unwanted disclosure of that information.

Why is confidential information in data a concern for foundation models?

If not developed in accordance with data protection rules and regulations, the model might expose confidential information or IP in the generated output or through an adversarial attack.

Parent topic: AI risk atlas

Generative AI search and answer
These answers are generated by a large language model in watsonx.ai based on content from the product documentation. Learn more