Skip to main content

Next Steps

📖 Lesson content

What you'll learn

Estimated time: 20 minutes

By the end of this lesson you'll be able to:

  • Synthesize the four properties and training fingerprints into a working mental model
  • Connect the Capabilities & Limitations framework to the 4D Framework as two halves of one system
  • Identify one concrete change to make in your AI practice this week

Applying the 4D framework to get better AI outputs

(5 minutes)

Fluent AI use isn't about memorizing every failure mode. It's about holding a small, clear model of the machine in your head, so that when something goes wrong you can recognize which kind of wrong it is and respond accordingly.

A small model of the machine

AI Capabilities & Limitations Framework

Four properties that shape what AI can and can't do for you. Each sits on a spectrum — the further right, the more you should verify and compensate.

Capability

Limitation

Next Token Prediction

Where do AI answers come from?

Well-worn paths: summarize, reformat, explain common conceptsNovel territory, sparse patterns, "true vs. sounds true"

Knowledge

What does AI actually know?

Frequent, recent-in-training, consistent: mainstream topics, popular languagesRare, post-cutoff, niche, local, or contested topics

Working Memory

What is the AI paying attention to right now?

Material fits comfortably, session is current, you supply relevant contextVery long docs/conversations, expecting cross-session continuity (the cliff)

Steerability

How much am I in control?

Short, concrete, verifiable instructions ("respond as a table," "under 100 words")Long reasoning chains, abstract asks, native precision

Key takeaways

  • You now hold a working mental model: four properties as continuums, characteristic failures as property intersections.
  • This framework and the 4D Framework are two sides of one system. The properties explain what the 4D competencies are responding to.
  • Calibrated trust means locating your task on each continuum and matching your verification and context habits to where it sits.
  • Models will keep changing. The shape of these properties stays useful even as the exact boundaries shift.

Exercises

Exercise: Your Commitment

Return one last time to your task list from Lesson 1. For each task, jot a quick gut-read: where does the task land on each property's continuum, and what mitigations might you need?

Now, pick one task and one change you'll make this week (a verification step, a standing-context setup, a checkpoint, a goal-stated-not-just-format habit). Write it down.

Lesson reflection

  • What's the single biggest shift in how you think about AI behavior from Lesson 1 to now?
  • Which of the 4Ds feels most immediately sharpened by what you've learned here?

What's next

If you haven't yet taken the AI Fluency Framework & Foundations course, that's the natural next step. It goes deep on the human competencies this course gave you the machine-side context for. And keep testing edges: the properties stay stable, but where the lines sit will keep moving as models improve.

Feedback

As you progress through the course, we'd love to hear from you about how you are using concepts from the course in your work, plus any feedback you may have. Share your feedback here.

🎬 Video transcript

Source video: F7ciHDKAlCA

📜 Click to expand transcript (cleaned + AI-translated)

Introduction: Building a Durable Mental Model for AI

When you started this series, you probably had some version of one question: "Why does AI do that?" But you're leaving with something better than a list of answers. You're leaving with a structure that lets you answer that question yourself.

Models change, features ship, and edge cases surprise people, but these core properties remain. What you've built in this course is the ability to stop being surprised.

The Two Stages of Model Training

Models are trained in two distinct stages:

  1. Pre-training: Builds a document completer.
  2. Fine-tuning: Layers an assistant on top.

Every behavior you see—whether helpful or strange—traces back to one of those two fingerprints.

The Four Core Properties of AI Models

Once models are created, they exhibit four primary properties:

  • Next Token Prediction
  • Knowledge
  • Working Memory
  • Steerability

Each of these exists on a continuum with a capability zone, a limitation zone, and product features that push the edge further out. The same mechanism is always running; the only thing that changes is where your specific task lands on that line.

Diagnosing Property Collisions

When something goes wrong, it's almost always the result of two of these properties colliding:

  • Fabricated Citations: This occurs when next token prediction meets a knowledge gap.
  • Conversation Drift: This happens when working memory fades while steerability causes the model to take new instructions too literally.

To adjust your approach, you must diagnose not just what broke, but which two properties collided.

Integrating the Four Properties with the 4D Framework

These four properties tie directly into the 4D framework: Delegation, Description, Discernment, and Diligence. These are not separate systems; the 4Ds are what you do, and the four properties are what you are responding to.

  • Discernment: Understanding next token prediction makes you better at discernment because you recognize that fluency and accuracy are independent variables.
  • Description: Understanding working memory improves your description because you know context is leverage and you stop assuming the model remembers everything.
  • Delegation: Understanding steerability sharpens your delegation because you know where control is high and where it is less precise.

The machine layer sharpens the human layer. They are opposite sides of the same coin.

Calibrated Trust as a Habit

Calibrated trust with AI is often discussed as an attitude, but it is actually a habit. Before handing a task to an AI, you should run a quick internal check:

  • Is this well-worn territory or sparse?
  • Is this topic recent or stable?
  • Is my context window comfortably inside the limit?
  • Are my instructions concrete, or is there room between my words and my intent?

Based on this check, you adjust your workflow:

  • More verification where fabrication is likely to concentrate.
  • More context where the model cannot guess your meaning.
  • More checkpoints when reasoning chains run too long.

You look for features that extend the capabilities of the models. You don't blindly trust or distrust the AI; you locate the task and set your habits accordingly. This is the essence of AI fluency.

Conclusion: Maintaining Fluency as Models Evolve

To continue growing, practice on real work. Your mental model of AI behavior gets sharper the more you test it against actual output. Pay attention to where your predictions are right and where they are off.

Revisit the 4D framework with these four properties in mind to gain a new lens on your workflow. Keep testing the edges. As models improve, context windows get bigger and features close existing gaps. While the specific numbers and the exact location of the "edges" will shift, the fundamental shape of these properties holds:

  • AI will remain a predictor whose fluency may exceed its accuracy.
  • It will continue to have uneven knowledge with specific cutoffs.
  • It will continue to work within a finite context window.
  • It will continue to follow instructions despite gaps between words and intent.

These facts do not expire when a new version is released. You have built a durable mental model that allows you to track the target even as it moves.

🔁 Related lessons

📚 Source & attribution

Was this lesson helpful?

Feedback / ReportSpotted an issue or have an improvement idea?