Prismo | Branches of AI (Part 2)
Search & Knowledge Representation: How AI Finds and Remembers
In last week’s article on Machine Learning and Logic-Based AI, we explored how some branches of AI learn from data while others follow rules. This week, we’re looking at two more branches: Search and Knowledge Representation, the parts of AI that deal with finding information and remembering it in a useful way.
AI in Plain Language
Search: Think of this as AI’s problem-solving GPS. Search algorithms explore possibilities step by step to find the best path to a solution. Just like Google Maps tests out different routes to see which is fastest, AI search methods test out options to solve puzzles, win games, or plan actions.
Knowledge Representation: This is AI’s memory system. It’s about storing information in a way that machines can use logically. Humans don’t just memorize facts — we connect them (“Paris is the capital of France,” “France is in Europe,” “Capitals are political centers”). Knowledge representation lets machines build similar connections.
Why It Matters for Professionals
Search shows up whenever software or apps suggest the “best option” — the shortest route, the lowest airfare, or the next chess move. It’s the decision-making backbone.
Knowledge Representation underlies tools that need structured reasoning, like chatbots in customer support or compliance checkers in finance. These systems need to remember relationships between facts to give consistent answers.
For professionals, these branches matter because they explain why some tools excel at planning and problem-solving, while others excel at structured Q&A.
Real-World Examples
Search in action:
Google Maps finding the fastest route.
Chess engines like Stockfish calculating millions of moves ahead.
Scheduling apps optimizing your calendar.
Knowledge Representation in action:
IBM’s Watson answering Jeopardy questions by linking facts.
Medical databases that connect symptoms to diagnoses.
Chatbots that rely on stored FAQs to provide consistent answers.
A Few Limitations to Keep in Mind
Search: Can be slow and resource-intensive when there are too many possibilities (the “combinatorial explosion” problem). For example, in chess, the number of possible move sequences grows astronomically with each turn. Even the fastest chess engines can’t explore every option, so they must prune the tree and approximate.
Knowledge Representation: Machines only “know” what’s explicitly encoded. If the knowledge base is incomplete or biased, the reasoning will be too. For example, a medical expert system trained on data from adults may fail to diagnose conditions in children, simply because that knowledge wasn’t encoded in the system.
Try It Yourself
Open an AI app like ChatGPT.
1. Ask it: “Find the shortest route between these cities: New York → Philadelphia → Washington D.C.” (You’ll see it mimic a search process.) You’ll likely get a step-by-step route (like a travel plan). Notice that the answer is about finding the best sequence of steps — that’s how search works.
2. Then ask: “What do New York, Philadelphia, and Washington D.C. have in common?” (It will retrieve structured knowledge about geography, history, or government, mimicking knowledge representation.) This time, the answer will list shared facts (all are East Coast cities, historically important, centers of government or culture). Notice that the answer is about connecting knowledge that’s already stored and related — that’s knowledge representation.
The difference isn’t in the style of writing, but in the type of reasoning: one is about finding paths forward (search), and the other is about connecting stored facts (knowledge representation).
Think About It
Where in your daily work do you depend on finding the best path (search), and where do you depend on recalling structured information (knowledge representation)?
Look for Search in Action:
When you’re planning your day — deciding which emails to tackle first, or which tasks to prioritize.
When you use a navigation app to find the fastest commute.
When your calendar app helps schedule overlapping meetings.
Look for Knowledge Representation in Action:
When you answer a client’s question by recalling company policies or product features.
When you fill out paperwork that depends on remembering exact categories (like tax forms).
When you explain how different parts of your business connect (“this department reports to that one,” “this tool integrates with that system.”)
Try noticing these moments today. You’ll start to see that your brain uses both search and knowledge representation constantly, and so do the AI tools being built to support you.
The Takeaway
AI isn’t just about learning — it’s also about searching smartly and remembering meaningfully. Search algorithms power the decision-making side, while knowledge representation helps machines build and use structured knowledge.
At Prismo: The 10-Minute AI School, we’re mapping out these branches so you can see how they all fit into the bigger picture of intelligence.
At Prismo: The 10-Minute AI School, we’ll continue breaking down these branches so you can see how they fit into your world of work.
💡 Curious about how this fits into your own career? Drop us a note — we love a good AI chat.
