According to the Microsoft Azure AI Fundamentals (AI-900) official study materials and Microsoft's Responsible AI Principles, one of the core principles is "Privacy and Security." This principle ensures that AI systems protect personal and sensitive information, maintaining compliance with privacy laws, data protection regulations, and ethical data-handling practices.
If a Natural Language Processing (NLP) model is created using data obtained without permission, it directly violates this principle. Data collected without proper consent breaches user privacy and potentially violates regulations such as GDPR (General Data Protection Regulation) or other global privacy frameworks.
The Privacy and Security principle emphasizes the following:
* AI systems must ensure data collection and usage transparency.
* Data must be lawfully acquired and used with consent.
* Systems should protect data against unauthorized access or misuse.
In contrast:
* Inclusiveness promotes accessibility and fairness for all users.
* Transparency focuses on explaining how AI systems make decisions.
* Reliability and safety ensure systems function as intended and minimize harm.
Therefore, using unapproved data clearly breaches Privacy and Security, as it involves unethical data sourcing and endangers user trust.