Role Prompting: Giving AI a Job

Think of an AI like a world-class method actor.  I mean eat your heart our Daniel Day Lewis.  If you just walk up to them on the street and ask for advice, they’ll give you a generic, polite answer.  But if you tell them, “You are a grumpy 1940s detective,” or “You are a gentle Read More …

Shot Prompting

Imagine you’ve just hired a brilliant, eager, but slightly literal-minded intern.  They have read every book in the library, but they have never worked in your specific field before. How you give them directions will determine whether they succeed or fail.  In the world of Artificial Intelligence (AI), these directions are called “prompts.”  The secret to Read More …

Multimodal Generative AI Concepts

Core Definition Multimodal AI is a type of artificial intelligence that can process, integrate, and reason across multiple types of data simultaneously. Unlike traditional AI that focuses on a single type of input, these systems attempt to fusion data to combine diverse sensory information into a unified understanding.  This approach is closer to human perception, Read More …

The AI Stack as seen by Google

This hierarchical view can be a great way to understand how the different components work together to deliver AI solutions within Google’s stack. 1. Infrastructure Layer The foundational infrastructure layer is the bedrock of the entire AI stack, serving as the essential engine that provides the physical and virtual resources necessary to power sophisticated AI Read More …

Grounding in AI

If you’ve ever used an AI, you might have noticed it can be a bit of a “know-it-all.”  It speaks with total confidence, even when it’s completely wrong.  This is where grounding comes in.  It’s the difference between a person guessing at an answer and a person looking it up in a specific group of Read More …