The Doing Got Cheap. Now What? | Claude Fable 5 Changes Work
Nate B Jones reviews the pulled Fable 5 model — which he believes is a ~10 trillion parameter pre-train — arguing its real significance is not being smarter but bigger: it carries whole jobs unattended, which exposes that most people's asks are still prompt-sized, and it demands a new skill he calls detailed task imagination plus a 'model manager' way of working.
AI News & Strategy Daily | Nate B Jones18 minTranscript found
Quick learning frame
Read this before watching.
Creative automation uses agents to accelerate production while keeping human taste in story, pacing, selection, and critique.
New playlist item from AI News & Strategy Daily | Nate B Jones; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to spot Fable-sized jobs — big, gnarly, untracked work — and package them as complete assignments with a data pack, a written definition of done, and owner-level review instead of prompt-sized asks.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01Brief
02Source
03Generation
04Selection
05Edit
06Taste Review
Deep lesson
Turn this video into working knowledge.
3,645 cleaned transcript words reviewed across 1,008 timed caption segments.
Thesis
The Doing Got Cheap. Now What? | Claude Fable 5 Changes Work teaches a practical creative automation move: Nate B Jones reviews the pulled Fable 5 model — which he believes is a ~10 trillion parameter pre-train — arguing its real significance is not being smarter but bigger: it carries whole jobs unattended, which exposes that most people's asks are still prompt-sized, and it demands a new skill he calls detailed task imagination plus a 'model manager' way of working.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:26
The big model feeling
“are dropping in the next month or so. You should be expecting this from open-source models in the next 4 5 6 months. I want to drop this Fable 5 model review now because, yes, I hope Fable...”
Fable 5 behaved unlike earlier models: instead of smoothing over garbage data it quarantined it, inventoried fake credentials without leaking them, and unprompted built a review queue of uncertain calls — behaving like it expected to be checked — letting Nate genuinely walk away from running work for the first time; Stripe reportedly compressed months of engineering into days, though it's expensive ($50 per million output tokens), has weak visual taste, and still ends every run with human review. List the failure modes you currently hover over AI to catch (invented sources, silent fixes, smoothed conflicts) and define what evidence would let you stop hovering on a delegated task.
8:54
Task imagination
“model. It's thoughtful, it's thorough, it tackles big task. It's exactly what I've been describing for code, right? It's something where you can ask it to refactor an entire repo and it can do it. So, you need...”
The gap between headlines and your workday is ask size: 2023–24 taught us to ask small, so every frontier model feels the same at prompt scale — the new skill is 'give, don't ask': assemble a pile of source material, a clear written paragraph of what done looks like, rough guidelines for judgment calls, then hand it over and walk away; spending 3–4 hours building the data pack is worth it if the job saves two weeks. Write down the 'weather' over your work — the gnarly jobs nobody owns that make you sigh — pick the most valuable one, and spend time locating the data you would need to hand the whole job to a model.
12:39
Become a model manager
“feeding. You need model managers for this model to do well. You need people who will be able to say, "This is the scope and scale and direction and this is the data that we're feeding this model...”
A model that does two weeks of work only kills pure-execution, zero-judgment jobs, because it needs heavy care and feeding: model managers who set scope, direction, and data, then review output like an owner reviewing a senior stakeholder's work — the people working with these models are working harder than ever, working themselves into new jobs, and an IC who does the exercise and ships Fable-scale wins is inviting a promotion, not a layoff. Reframe your role in one paragraph as a model manager: for your biggest current project, specify the scope, the data the model needs, the judgment calls it can make alone, and how you will review the finished work.
01
Brief
Start with this video's job: Nate B Jones reviews the pulled Fable 5 model — which he believes is a ~10 trillion parameter pre-train — arguing its real significance is not being smarter but bigger: it carries whole jobs unattended, which exposes that most people's asks are still prompt-sized, and it demands a new skill he calls detailed task imagination plus a 'model manager' way of working. Treat "Brief" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:26, where the video says: “are dropping in the next month or so. You should be expecting this from open-source models in the next 4 5 6 months. I want to drop this Fable 5 model review now because, yes, I hope Fable...”
02
Source
Use "Source" to locate the part of the creative automation workflow the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 8:54, where the video says: “model. It's thoughtful, it's thorough, it tackles big task. It's exactly what I've been describing for code, right? It's something where you can ask it to refactor an entire repo and it can do it. So, you need...”
03
Generation
Turn "Generation" into the reusable artifact for this lesson: A creative workflow board with critique criteria and review checkpoints. This is where watching becomes something you can inspect and reuse.
04
Selection
Use "Selection" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Edit
Use "Edit" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Taste Review
Use "Taste Review" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
Example
Source-backed work packet
Convert the video into a scoped task that includes the transcript claim, target workflow, acceptance criteria, and proof. The output should be a creative workflow board with critique criteria and review checkpoints..
Example
Claim vs. demo brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the workflow.
Example
Teach-back module
Transform the lesson into a definition, a mechanism diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
Letting the prompt drift into generic advice that could apply to any video in the playlist.
Copying the tool setup without identifying the operating principle that transfers to your own stack.
Skipping the artifact, which means the learning never becomes operational or inspectable.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: Nate B Jones reviews the pulled Fable 5 model — which he believes is a ~10 trillion parameter pre-train — arguing its real significance is not being smarter but bigger: it carries whole jobs unattended, which exposes that most people's asks are still prompt-sized, and it demands a new skill he calls detailed task imagination plus a 'model manager' way of working.
02
Explain the practical stakes without hype: New playlist item from AI News & Strategy Daily | Nate B Jones; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the Brief -> Source -> Generation -> Selection -> Edit -> Taste Review sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A creative workflow board with critique criteria and review checkpoints.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: The Doing Got Cheap. Now What? | Claude Fable 5 Changes Work
- URL: https://www.youtube.com/watch?v=2w_vwQVvFmc
- Topic: Creative Automation
- My current learning frame: Choose one painful, untracked, two-week-sized job in your work, spend a few hours assembling its data pack and a one-paragraph definition of done, hand the whole thing to the most capable model you can access, walk away, and then review the output like an owner — logging what the model surfaced for your judgment.
- Why this matters: New playlist item from AI News & Strategy Daily | Nate B Jones; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:26 / Evidence 1: "are dropping in the next month or so. You should be expecting this from open-source models in the next 4 5 6 months. I want to drop this Fable 5 model review now because, yes, I hope Fable..."
- 2:45 / Evidence 2: "the data instead of fixing it. It found the fake credentials and inventory them without leaking them and then, and this is the part that got me, it built me a review queue. Every call it wasn't sure..."
- 4:44 / Evidence 3: "that's the kind of scale that you want to give this model. Think back. In 2023 and 2024, asking big got you burned, right? You handed a model something real, and it lost the thread by step six,..."
- 7:19 / Evidence 4: "This is about jobs that are bigger than that, that aren't on anybody's tracker yet because they're dirty and ambiguous. And yes, I'm picking those big numbers on purpose because these are numbers that you need to make..."
- 8:54 / Evidence 5: "model. It's thoughtful, it's thorough, it tackles big task. It's exactly what I've been describing for code, right? It's something where you can ask it to refactor an entire repo and it can do it. So, you need..."
- 12:39 / Evidence 6: "feeding. You need model managers for this model to do well. You need people who will be able to say, "This is the scope and scale and direction and this is the data that we're feeding this model..."
- 17:14 / Evidence 7: "I've got a a set of Fable specific skills that help when Fable is struggling with something for you. For example, writing. Like if you need to get Fable to read your voice, what does that look like?"
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable claims from the video. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A creative workflow board with critique criteria and review checkpoints.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: Brief -> Source -> Generation -> Selection -> Edit -> Taste Review
- 3 concrete examples that apply the video idea to real agentic work
- 2 failure modes the video helps prevent
- a checklist I can use the next time I run Codex or Claude
- one practical exercise with a clear done signal
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "The Doing Got Cheap. Now What? | Claude Fable 5 Changes Work", not a generic Creative Automation essay.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- If evidence is weak, say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
Creative AI removes the need for taste.
It increases the need for taste because output volume explodes.
The best prompt is enough.
References, critique, iteration, and post-production matter just as much.
Practice studio
Learning only counts when you make something.
01
Transcript evidence map
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a creative workflow board with critique criteria and review checkpoints..
A reusable artifact with a done signal and one verification step.03
Teach-back card
Explain the lesson to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
What specific behavior with messy data convinced Nate that Fable 5 felt categorically bigger?
How does 'detailed task imagination' differ from ordinary delegation?
Which jobs does Nate say a model like Fable 5 will actually eliminate, and why not more?
Source shelf
Use the video as a doorway, then verify with primary sources.
