Skip to content

AI

Terms

Principiles

There are six principles of responsible AI:

  • Fairness: Fairness focuses on preventing bias and ensuring equitable treatment for all individuals. It involves designing AI systems that do not discriminate against any group. In this scenario, there is no mention of addressing bias or ensuring equal outcomes, so fairness is not the principle being applied.
  • Reliability and safety: Reliability and safety are about ensuring that AI systems perform consistently and safely under expected conditions. This includes rigorous testing and validation to avoid harm. The question does not refer to testing or operational safety; it only talks about sharing information, so this principle does not apply here.
  • Privacy and security: Privacy and security involve protecting user data and ensuring that the system is secure against threats. This principle is critical when handling sensitive information. However, the scenario does not mention data protection or security measures, so it is not relevant in this case.
  • Transparency: The principle of Transparency helps people understand how to use AI solutions, including their behavior, possibilities, and limitations.
  • Inclusiveness: Inclusiveness means making AI accessible and beneficial to everyone, regardless of background or ability. It focuses on avoiding discrimination and promoting equal opportunity. Since the question does not mention accessibility or inclusion for diverse users, this principle is not being addressed.
  • Accountability: Accountability ensures that there is clear responsibility for AI decisions and outcomes. It involves governance and mechanisms to hold individuals or organizations answerable for the system’s behavior. The scenario does not mention assigning responsibility or governance, so accountability is not the principle applied here.

Workkloads

AI:

  • Anomaly Detection: This is a technique used within Machine Learning to identify unusual patterns. It’s not a foundational element of AI itself.
  • Object Detection: This is a specific application within Computer Vision. It’s not a standalone key element of AI.
  • Knowledge Mining: Although Microsoft uses this term in solution areas, it’s not one of the five foundational elements of AI. It combines multiple services like search and cognitive skills, but isn’t a core AI domain like ML or NLP.

Other:

  • Machine Learning: This is the core foundation of AI. It enables systems to learn from data and improve over time without being explicitly programmed.
  • Computer Vision: A key AI domain that allows machines to interpret and understand visual information like images and videos.
  • Natural Language Processing: This enables machines to understand, interpret, and generate human language, making it essential for tasks like translation, sentiment analysis, and chatbots.
  • Automated Machine Learning: While it’s a tool or approach within Machine Learning, Microsoft includes it as part of its AI strategy to simplify and scale ML model development.