Understanding AI and Data
What is AI
Artificial Intelligence
What is Algorithmic Bias?
This bias occurs when an AI system reflects unfair patterns found in its training data.
What is personally identifiable information (PII)?
Data that can be reasonably linked to an identifiable individual, either directly or indirectly.
What is differential privacy?
This privacy‑preserving technique adds statistical noise to protect individual data points while allowing aggregate analysis.
What is personal data?
Information that can identify an individual directly or indirectly and is often used to train AI systems.
What is personally identifiable information (PII)?
This kind of data can identify an individual, such as a name or student ID.
Who is responsible when an AI system makes a harmful mistake—the developer, user, or company?
No single party is held fully responsible.
What is data leakage?
This privacy risk arises when AI models unintentionally memorize and reproduce sensitive training data.
What is HIPAA?
This U.S. sector‑specific law governs the privacy and security of medical data often used in healthcare AI systems.
What is sensitive personal data?
This type of personal data includes health records, biometric data, and racial or ethnic information.
What is a training dataset?
AI tools are trained on this kind of data set to recognize patterns.
Can AI ever truly be unbiased, or will it always reflect human biases?
AI is inherently shaped by human society
What is informed consent?
This concept refers to the legal and ethical requirement that individuals understand and agree to how their data will be used before it is collected.
What is AI accountability?
the principle that humans and organizations can be clearly identified, assigned responsibility, and held legally and ethically answerable for the actions and outcomes of AI systems.
What is the right to be forgotten?
This legal right allows individuals to request deletion of their personal data from AI systems and databases.
What is data leakage or algorithmic bias?
This term describes how AI can unintentionally reveal sensitive or biased information.
What is encryption?
It's a process that ensures that data sent between systems cannot be easily read by unauthorized users.
What is a membership inference attack?
This machine‑learning privacy attack attempts to determine whether a specific individual’s data was included in a training dataset.
What is purpose limitation?
This privacy principle limits data use to the original reason it was collected, even when training AI models.
What is federated learning?
This method allows AI models to be trained across multiple devices without centralizing users’ personal data.
What is output data or generated content?
Educators should understand the difference between input data and this kind of data.
What is data misuse
This privacy risk happens when AI systems are trained on sensitive data without user permission.
What is purpose limitation?
This legal principle restricts the use of personal data to the original reason it was collected, even when repurposed for AI training.
What is the GDPR (General Data Protection Regulation)?
This major EU regulation governs how personal data may be collected, processed, and used—including by AI systems.
What is balancing data privacy and AI innovation?
This ongoing challenge in AI governance focuses on limiting harm while still enabling innovation using personal data.