AI Terminology
Machine Learning (ML) - A way for computers to learn from data and improve over time without being explicitly programmed. This is probably one of the most important breakthroughs in modern AI as it opens the doors for AI-augmented tools to draw on all the knowledge we can feed it without having to program and re-program the underlying databases. This is also arguably what makes AI appear to be "intelligent" as it can keep refining its analytical outputs as more and more data is introduced. It also leads to the concept of self-learning in that the output of an AI session can be fed back into its underlying data set to effectively add its own "intelligence" to augment what is provided by the programmers.
Computer Vision - AI that allows machines to interpret and analyze visual information, such as recognizing objects in images. This type of AI, while greatly enhanced with the sophisticated data algorithms used, finding objects in images has been part of software like Photoshop for a long time. What is brings to AI, is the ability to interact with the world in much the same way as we do, by seeing things. As a simple but important application, border security requires real-time processing of the imaging available at the various border crossings. This means more than just face recognition, which has been available for a long time, but also now the ability to recognize dangerous or suspicious body movements. An interesting application would be watching for pick-pockets in crowded airports.
Natural Language Processing (NLP) - AI's ability to understand, interpret, and generate human language. Unlike programming languages, which are designed with unambiguous syntax and semantics, spoken languages are far from having these attributes. It was very important to the builders of today's AI systems to be able to conquer this challenge. In doing so, platforms like ChatGPT could be created that can interact with human users on most any communication approach desired. This mostly means the underlying data models for the languages are filled with critical mappings for style usage, spelling, venacular, sentence structure, presentation approaches and much more. This is one of the areas you can safely explore with tools like ChatGPT from the start. Have some fun - "Can I ask you questions using pig latin as the input? Yes, you can ask questions in Pig Latin, and I’ll do my best to interpret and respond! If your question is complex or unclear, I may ask for clarification in standard English to ensure accuracy. Feel free to try it out!"
More to come ...