From rules to agents · Learn visually

Interactive Illustrated AI History

Explore the evolution of AI from rules and statistical learning to deep learning, large models, RAG, and agents through clickable, step-by-step teaching demos.

Start with the overview Browse all chapters

Guided causal stories

Start with 3 questions across AI history

Begin with a core question and follow a reviewed sequence to see what each step inherited, solved, and left open.

6 guided steps

From Experience Updates To Feedback Loops

How did outcomes after actions enter value learning, deep representations, preference post-training, and agent systems?

Open timeline View focused lineage

7 guided steps

From Rules To Learned Representations

When search, rules, uncertainty modeling, and hand-crafted features each met their limits, how did AI gradually shift toward data-driven and learnable representations?

Open timeline View focused lineage

7 guided steps

From Scaled Models To Reliable Systems

When scaled language models enter concrete applications, why might they still need external knowledge, post-training, action loops, and ongoing risk evaluation?

Open timeline View focused lineage

Recommended learning path

Follow 13 chapters to see how AI evolved step by step

About 5 minutes · Read the main thread

Use one learning map to connect the evolution from rules and statistical learning to deep learning, RAG, and agents.

Demo 01Search Trees / A*

Switch among BFS, DFS, and A* to see how search strategies affect the frontier.

Demo 02Expert System Rule Reasoning

Select conditions and add exceptions to see how if-then rules produce conflicts.

Demo 03Bayesian Updating

Adjust the prior and evidence strength to see how evidence updates belief.

Demo 04Decision Boundaries

Compare linear, nonlinear, and overfit boundaries to understand data-driven learning.

Demo 05CNN Kernels

Choose a kernel and advance the window to see how a feature map is produced.

Demo 06Reinforcement Learning And Feedback

Compare immediate and delayed rewards, then separate training updates from runtime observations.

Demo 07Attention

Select a token and compare direct Attention connections with RNN chain propagation.

Demo 08Foundation Model Lifecycle

Separate what pretraining, instruction tuning, preference feedback, and runtime context change.

Demo 09LLM System Map

Understand why context, retrieval, tools, memory, and evaluation surround large models.

Demo 10RAG Pipeline

Follow a question through embedding, retrieval, reranking, prompting, the LLM, and a cited answer.

Demo 11Agent Loop

Run the loop of planning, tool calls, observation, revision, and a final answer.

Demo 12Safety / Eval

Compare normal and risky requests, save a failure as a regression test, and run the release gate.

Explore Further

Revisit AI's evolution from different perspectives

TimelineAI Evolution Timeline

Connect search, rules, statistical learning, deep learning, Transformer, RAG, and agents across eras.

LineageAI Technical Lineage

See how symbolic AI, statistical learning, neural networks, foundation models, and agents relate by paradigm.

DiagramsReusable Diagrams And Exports

Browse downloadable diagrams and key views designed for sharing and review.