Preparing Practice Exam…
Please wait a moment while the system generates the exam questions, answers, and study session.
Please wait a moment while the system generates the exam questions, answers, and study session.
ISACA Advanced in AI Security Management (AAISM) · Isaca
Select one answer.
Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.
The primary legal risk when generative AI accesses and uses internet data is intellectual property (IP) and copyright infringement. Copyright holders, artists, and publishers frequently sue AI providers for using their protected work in training datasets or generating outputs that mimic copyrighted material without permission.
The primary source of lawsuits against generative AI service providers (e.g., OpenAI, Anthropic, Meta, Stability AI) stems from unauthorized use of copyrighted internet-sourced data for model training, leading to claims of direct infringement. Activating proactive filtering logic (e.g., content moderation, opt-out mechanisms, or exclusion based on known IP/copyright indicators like robots.txt flags, known copyrighted works, or metadata) directly prevents ingestion of high-risk data, reducing exposure at the source. This is a technical, preventive control that aligns with best practices for mitigating IP-related litigation risks in generative AI.