You use it and cannot explain it
The answers arrive and you have no model of where they come from.
Courses
Start with three things you can do today
How does AI actually work, explained without jargon
Know what modern AI can and cannot do, and why it gets things wrong.
AI is a broad category. A language model generates text using learned patterns and its current context; the surrounding product may also retrieve information or use tools. Fluent text can still be wrong. The example below separates an answer, its evidence and the system that produced it.
Want new features first?
A starting goal
I want to understand what modern AI can and cannot do. Explain how it learns patterns, why it makes mistakes, and how to judge claims about its capabilities.
Change any word so it fits your situation. The course is built from what you type and a few setup questions.
Start today
Three things you can do before the course exists.
Useful on their own. The course is for when you want the whole skill, not just the first move.
-
Separate the model from the whole product
Original hypothetical example: you ask, "When does the museum open today?" A product might generate an answer from learned information, retrieve the museum website, or report that it cannot check. Those are different routes to an answer, even if each produces the same sentence.
-
Inspect evidence, not just wording
If the answer says "10am," ask what supports that time. Open the museum's current official page and check the date and holiday exceptions. A supplied citation is a lead to inspect, not a guarantee that the page supports the sentence.
-
Explain the distinction back
A useful explanation names the model, the information available in context and any observed tool use. It does not call the model a complete copy of everything it read or infer live search from a confident tone. This illustration is not a test of a particular provider.
Sources: NIST: generative AI errors, evidence and privacy risks. Practical examples are starting points, not promised results.
- Best for
- AI & technology
- Prerequisites
- None. Bring the situation you have.
- Course length
- 7 to 30 lessons
- Pace
- Yours. One lesson at a time, as often as you like.
- Format
- Read it, or play the audio
- Access
- Available on the App Store and Google Play, with an optional early testing list
Before the course is built
Setup asks about your situation first.
Lesson 1 opens on the moment you describe, and the examples use your role. For this request, expect questions like these.
- Do you use AI tools already, and for what?
- Is your interest practical, professional, or mainly curiosity about how it works?
- Which claim about AI do you most want to be able to judge?
Right for now
If any of this sounds familiar, start here.
The hype and the fear both sound plausible
You cannot tell which predictions deserve attention.
You are asked for an opinion at work
Someone wants to know what AI means for your team and you would like your answer to be grounded.
Course arc
Where the course goes, in order.
Cadence writes the lessons for you, so the exact path follows your answers. This is the ground a course built from this sentence covers.
01 / 06Learning from patterns
How training changes a model using data, and how this differs from looking up a source.
Why it sounds so sure
Fluency versus accuracy, and where confident nonsense comes from.
What it is good at
How to inspect demonstrated capability on a defined task.
What it is bad at
Checking errors on a particular model and task without assuming a permanent capability ranking.
Reading the claims
Benchmarks, demos and predictions, and how to weigh each.
Your own working model
A short account of how AI works that you can explain to someone else.
What changes
What you will be able to do, and what you will stop believing.
You will be able to
Explain how models learn
Describe pattern learning from data in plain words, without math.
Predict where it fails
Recognize why fluent output can still require factual checking.
Judge capability claims
Ask the questions that separate a demonstrated capability from a projected one.
Ideas this course corrects
Fluency proves that every fact was checked
A system can produce confident errors. Check the evidence needed for the decision.
All AI systems have the same strengths and limits
Capabilities depend on the model, tools, task and evaluation conditions. Broad claims need current evidence.
Course questions
What to know before you start.
How does AI actually work, explained without jargon: what can I do today?
Separate the model from the whole product. Original hypothetical example: you ask, "When does the museum open today?" A product might generate an answer from learned information, retrieve the museum website, or report that it cannot check. Those are different routes to an answer, even if each produces the same sentence. Inspect evidence, not just wording. If the answer says "10am," ask what supports that time. Open the museum's current official page and check the date and holiday exceptions. A supplied citation is a lead to inspect, not a guarantee that the page supports the sentence. Explain the distinction back. A useful explanation names the model, the information available in context and any observed tool use. It does not call the model a complete copy of everything it read or infer live search from a confident tone. This illustration is not a test of a particular provider.
Is this a fixed course?
No. Cadence builds the course when you type the sentence and answer a few setup questions, so the lessons follow your situation. This page describes what a course built from this request covers. You see the whole path before Lesson 1.
Who is this course for?
AI is a broad category. A language model generates text using learned patterns and its current context; the surrounding product may also retrieve information or use tools. Fluent text can still be wrong. The example below separates an answer, its evidence and the system that produced it. It sits under AI & technology in the Work & ambitions part of the library.
Do I need any prior experience?
No. Setup asks what you already know and where you want to use it, and the course starts from there. Bring a real situation if you have one; the examples are built around it.
How long does it take?
Most courses run 7 to 30 lessons. The course length is sized to what you asked for. You set the pace: one lesson at a time, as often as you like. Lessons can be read or played as audio.
Does this course award a certificate or degree?
No. Cadence is for practical self-learning and does not award accredited degrees or professional certification.
Start today
Type it, answer a few questions, open Lesson 1.
A starting goal
I want to understand what modern AI can and cannot do. Explain how it learns patterns, why it makes mistakes, and how to judge claims about its capabilities.
Want new features first?