Anthropic shipped a free AI Fluency for Small Business course with PayPal on May 13, alongside the Claude for Small Business launch and the start of a 10-city in-person SMB tour. Most of the coverage I have seen treats the course as a footnote — “and there is also a free training” — which sells it short. The course is actually a coherent 54 minutes of video, structured around a framework that is genuinely useful, with two real small-business owners doing the demos.
Here is the lesson-by-lesson breakdown so you can decide whether to invest the hour, and what to do with it afterward.
The basic facts
- Title: AI Fluency for Small Businesses
- Format: 9 video lessons + 1 quiz + a certificate of completion
- Total video: 0.9 hours — call it 54 minutes
- Cost: Free, ungated, no Anthropic subscription required
- Where it lives: anthropic.skilljar.com (Anthropic’s official course platform)
- Recommended prerequisite: AI Fluency: Framework & Foundations, the broader free course Anthropic shipped earlier. Skip if you are short on time — this SMB version is self-contained.
- What you need to follow along: access to any chat AI (Claude.ai is the default, ChatGPT or Gemini also work)
The course is built around the 4D Framework: Delegation, Description, Discernment, and Diligence. Those are the four skills you need to use AI well in a business setting, and the course teaches them in pairs over the four sections.
Section 1: Introduction and the AI Fluency framework (2 lessons)
This is the conceptual setup. The first two lessons define AI Fluency, walk through the 4D Framework, and ask you to write down your business’s values, goals, and constraints so every future AI interaction starts from the right context.
Lesson 1: What AI Fluency means. The pitch is that AI is a collaborator, not a tool, and being fluent means being able to direct that collaboration cleanly. The 4Ds get introduced as the four interconnected competencies you will keep coming back to.
Lesson 2: Articulating your values, goals, and constraints. This is the lesson everyone tries to skip and then regrets. The exercise is to write three short paragraphs — what your business stands for, what you are trying to achieve in the next 12 months, and what trade-offs are off the table. Without this written down, every later prompt feels generic. With it written down, you can paste it into the chat at the top of every session and Claude (or whichever AI you use) suddenly sounds like it understands your business.
Worth watching: Yes for everyone. Even if you have used Claude for six months, the second lesson’s writing exercise produces something you will reuse for years.
Section 2: How AI works (2 lessons)
This is the part that makes the course better than 90% of corporate AI training. Most courses skip the mechanics; this one builds you a working mental model so you understand why AI fails the way it does.
Lesson 3: A working mental model of today’s AI. Plain English on what large language models actually are and what they reliably can and cannot do. The framing is that AI is fluent but not omniscient — it can talk about anything but is most reliable on things it can verify, least reliable on things that require fresh real-world data.
Lesson 4: The next-token simulator. This is the standout lesson. Anthropic embedded an interactive simulator that walks you through how the AI picks the next word in a sentence based on probabilities. Once you have played with it for three minutes, you understand viscerally why prompts that give more context produce better answers, and why AI sometimes confidently invents details. It is the closest thing to “showing instead of telling” that any course in this category has done.
Worth watching: Yes for everyone. The next-token simulator is genuinely worth the click-through.
Section 3: Using AI in practice (2 lessons)
This is where the 4D Framework becomes a workflow. The two lessons pair the four competencies into two loops you actually run.
Lesson 5: The Description-Discernment loop on research. Description means writing prompts with enough context that the AI knows what you want. Discernment means critically evaluating what the AI produces. The loop pairs them: you describe your research question well, you discern whether the answer is trustworthy, you describe better, you check again. The demo uses a small-business research example — picking a new supplier — and you can see the prompt iterate from generic to specific over the course of the lesson.
Lesson 6: The Delegation-Diligence loop on customer data. Delegation means deciding which tasks to hand over to AI. Diligence means responsibly handling the outputs, especially when they touch customer data. The demo is built around a real customer-data task — segmenting a list, drafting follow-ups — and the lesson is careful about the parts you cannot delegate. This is the lesson PayPal’s involvement shines through; the responsibility framing is concrete in a way most AI training is not.
Worth watching: Yes if you have any sales, customer support, or marketing function. The Delegation-Diligence loop is the framework you will quote internally for a year.
Section 4: Putting it all together (3 lessons)
The final three lessons are the practice ones — where you take the framework and produce something for your own business.
Lesson 7: Building a repeatable AI-augmented workflow. You pick one task you do regularly (writing a weekly customer email, building monthly sales reports, drafting hiring posts) and turn it into a written workflow with explicit Description and Diligence steps. The course gives you a one-page template. The output is one workflow you can hand to a new hire on day one.
Lesson 8: Writing a short, honest AI use policy. This is the lesson most small business owners would never write on their own and need to. The template is intentionally short — about ten lines covering what AI is used for in your business, what data does not go in, who is responsible for double-checking, and how customers are told about AI involvement. It is the kind of document a board or a lawyer would otherwise charge you to draft.
Lesson 9: Planning next steps and earning your certificate. You complete a brief quiz, write down three commitments for the next 30 days, and download a shareable certificate.
Worth watching: Yes for the policy lesson alone. The certificate is a bonus, and yes, you can post it to LinkedIn.
The two small businesses you will see throughout
The course features real owners from two businesses, and they make the demos feel less abstract than usual.
Prospect Butcher Company (Brooklyn, NY). Corey Hammond runs a butcher shop and uses AI for inventory planning, supplier research, and customer email. His segments are good on the practical “we cannot afford to get this wrong” framing.
MAKS TIPM Rebuilders (California). Mak Cabessa runs an automotive power-module rebuilder — a small specialist shop. He uses AI for technical service writing and customer estimates. His demos are great on the “I am not a tech person” angle.
Both businesses are doing actual SMB work, not Silicon Valley examples. That is the choice that makes this course feel different from most corporate AI training.
What this means for you
Five concrete profiles where the course earns its 54 minutes.
If you have never used AI seriously and you run a small business: Watch all 9 lessons in order. Block out 90 minutes total — 54 of video plus 36 of pausing to write things down in the exercises. The order matters. Do not skip ahead.
If you have been using ChatGPT for a year but never wrote down what you use it for: Watch lessons 2, 4, 7, and 8 — about 30 minutes of video. The values-and-goals exercise (Lesson 2), the next-token simulator (Lesson 4), the workflow builder (Lesson 7), and the AI use policy (Lesson 8) are the high-leverage ones for you.
If you manage a team and want to set internal AI policy: Watch Lesson 8 first, then 7, then 2. Use the policy template from Lesson 8 as a starting point for a team document. This is the lesson worth getting your operations lead to also watch.
If you are on the 10-city SMB Tour list and trying to decide whether to attend: The free workshop covers the same material as the course plus a hands-on session in-person. If you can swing the half-day, the in-person version is better because you ask real questions in real time. If you cannot, the course covers 80% of what the workshop teaches. Either way, finish the course before you go — it sets up the workshop conversations to be deeper.
If you have an accountant, bookkeeper, or part-time admin who interacts with AI on your behalf: Send them the course link. The Delegation-Diligence loop (Lesson 6) is the lesson they need most, and the AI use policy (Lesson 8) is the document they should help write.
What this can’t fix
Four honest limits before you commit the hour.
- It does not teach you Claude specifically. Examples use Claude.ai but the course is about the framework, not the product. If you want a Claude-specific walkthrough — connectors, dispatch tasks, plugins — you will need a different resource.
- There is no spreadsheet, no QuickBooks, no HubSpot demo. The two SMB owners do their work; you do not get to watch them connect specific apps. If you wanted “set up your QuickBooks connector in 5 minutes” content, this is not it.
- The 4D Framework is genuinely good but it is also abstract. You will get more out of the course if you bring a real, current business task to the workflow lesson (Lesson 7) than if you sit through it passively.
- The quiz is light. Twelve questions, mostly comprehension. Do not skip it — it is worth doing the certificate — but do not expect it to test your real fluency. That happens in the work after.
The bottom line
It is free, it is 54 minutes, it produces three artifacts you will reuse for years (your values-and-goals paragraph, your AI workflow template, your AI use policy), and it is structured around a framework that holds up. The hour spent here is among the cheapest investments you can make in your small business this quarter.
After the course, the next step is putting the framework against real work — which is where our Claude for Small Business and AI Business Automation courses pick up. They walk through specific workflows (invoice chasing, monthly close, campaign runner) on top of the framework Anthropic teaches.
Sources
- AI Fluency for Small Businesses — Anthropic Academy
- PayPal partners with Anthropic to Close the AI Gap for Small Businesses — PayPal newsroom
- PayPal partners with Anthropic on launch of free AI Fluency for Small Business online course — EdTech Innovation Hub
- PayPal and Anthropic Take AI to Main Street Businesses — PYMNTS
- Introducing Claude for Small Business — Anthropic
- AI Fluency: Framework & Foundations — Anthropic Academy (the broader prerequisite)
