Home/ Tool index/ ELIZA Simulator
Launch ready Session 1 · Text Interactive

ELIZA Simulator

A rule-based chatbot made inspectable. It shows how pattern matching, templates, and conversational polish can feel meaningful without any language-model understanding — making the gap between visible mechanism and felt effect the thing to study.

tools/eliza-simulator/

Live preview · launch for the interactive version

§ A · What it makes visible

what the screen literally shows
Fig. 01

Rule matching

Every ELIZA response traces back to a pattern: a keyword, a template, and a fill-in. The rule that fired is right there on screen — before the response, not after.

Fig. 02

Implicit patterns

Even with the rules visible, the conversation can feel caring or knowing. That gap — between what the mechanism does and what the output implies — is the central thing to name.

Fig. 03

LLM contrast

ELIZA’s rules show what a modern language model hides. When you compare them side by side, the opacity of the LLM becomes visible by contrast, not by description.

§ B · How to investigate it

run it like an experiment, not a toy

Don’t just chat with it — run it like an experiment. Predict what rule will fire, then check.

01 · Predict

Before you type

Write down what you expect ELIZA to do with a specific input. Which keyword will it match? Which template will fire?

prompt: “I feel like nobody listens”
02 · Change one thing

Remove or change the keyword

Add, remove, or rephrase the triggering keyword. Does the matched rule change? Does the response change?

“feel” → “think” · same concern, different word
03 · Compare evidence

ELIZA vs. LLM

Run the same input through ELIZA and a pre-generated LLM example. What does each response reveal about what the system understood?

identical input · two very different mechanisms
04 · Name the gap

Not “ELIZA is fake”

Name the specific gap: what does the visible rule not do that the LLM response appears to do? The LLM’s gap is just harder to see.

felt understanding · vs. · demonstrated understanding

§ C · Debrief questions

after the investigation
What did you notice when you could see the rule that fired?
When did ELIZA feel meaningful even though the mechanism was visible?
What does ELIZA make visible that a modern LLM hides?
Where could the same gap appear in an AI product you use regularly?