13 · Anthropic's own training material

What Anthropic publishes to teach its own products, what's worth assigning, what to avoid putting in front of executives — and the substantial gaps you will have to fill yourself.

The headline: /learn is a router, not a curriculum

anthropic.com/learn is branded "Anthropic Academy", with the subhead: "Get in the know with Anthropic resources. From API development guides to enterprise deployment best practices, the academy has you covered."

It contains seven items: three persona tiles, two featured courses, a link to the course platform, and a newsletter signup. That is the entire hub.

The thing to know before you plan around it

The genuinely valuable executive material is not on /learn and is not linked from it. The hub omits the course catalogue with durations, the Cowork deployment guide and its maturity model, the role-specific Cowork tutorials, the customer stories, and the communications kits. Build from /learn alone and you will produce a developer-flavoured curriculum and miss most of what you need.

Where the material actually lives

What you wantWhere it actually isLinked from /learn?
Course catalogue with durations and categoriesclaude.com/resources/courses
Adoption maturity model & rollout roadmapThe Cowork deployment PDF — see §12
Role-specific Cowork walkthroughsclaude.com/resources/tutorials
Customer stories with figuresclaude.com/customers
Rollout comms templatesClaude Code communications kit — see §12
Courses (no durations, no categories)anthropic.skilljar.com
Business-oriented link hub/learn/claude-for-work
Developer link hub/learn/build-with-claude
Personal / AI Fluency hub/learn/claude-for-you

The course catalogue

All courses are free, hosted on Skilljar under Anthropic branding, require an account signup, and most award a certificate of completion. Verified directly against claude.com/resources/courses on 2 August 2026.

Professional — the executive-relevant track

CourseDurationContentExecutive fit
Claude 1011 hr12 lectures✓ The clean on-ramp
Introduction to Claude Coworknot published5 modules, ~17 lessons + quiz✓ The only Cowork course
AI fluency: Framework and foundations1.1 hr14 lectures, 1 quiz✓ Role-agnostic, framework-led
AI Fluency for Builders1 hr10 lectures, 1 quiz◐ "Builders" skews technical
AI Fluency for Small Businesses0.9 hr9 lectures, 1 quiz◐ Names "leadership roles"; SMB framing
AI Fluency for pK-12 Educators1 hr8 lectures, 1 quiz✗ Wrong audience

Engineering — do not assign to executives

CourseDurationContent
Claude Code 1011 hr12 lectures
Claude Code in Action1 hr15 lectures, 1 quiz
Introduction to subagents20 mins4 lectures
Introduction to agent skills30 min6 lectures
Claude Platform 1011 hr12 lectures, 1 quiz
Introduction to Model Context Protocol1 hr16 lectures, 1 quiz
Model Context Protocol: Advanced Topics1.1 hr15 lectures, 2 quizzes
Building with the Claude API8.1 hr84 lectures, 10 quizzes
Claude with Amazon Bedrock8 hr85 lectures, 10 quizzes
Claude with Google Cloud's Vertex AI8 hr85 lectures, 10 quizzes

Education & Nonprofits

CourseCategoryDurationContent
AI capabilities and limitationsEducation15 min13 lectures, 1 quiz
AI Fluency for EducatorsEducation24 min4 lectures
AI Fluency for StudentsEducation30 min5 lectures
Teaching AI FluencyEducation36 min7 lectures, 1 quiz
AI Fluency for nonprofitsNonprofits54 min9 lectures, 1 quiz
The one to steal from another category

AI capabilities and limitations (15 min, Education) is the best short literacy module Anthropic publishes for a leadership audience, despite being filed under Education. If you have fifteen minutes of pre-work to assign before an executive session, use this.

The three courses worth assigning

1 · Claude 101 — the executive on-ramp

Modules: Meet Claude (what it is, first conversation, getting better results, working on desktop) · Organizing your work and knowledge (projects, artifacts, skills) · Expanding Claude's reach (connecting tools, enterprise search, research) · Putting it all together (use cases by role) · Conclusion & certificate.

2 · Introduction to Claude Cowork — the only Cowork course

Anthropic's description, verbatim: "Cowork is Claude working directly with your files, folders, and apps — reading, editing, and producing real outputs on your machine."

ModuleCoversMaps to
1 · Meet Claude CoworkWhat it is, setup, what it can do, first task§02, §04
2 · Make Claude Cowork yoursBetter results; standing context (global instructions and projects); skills; "plugins: encode your team's expertise"§06
3 · Use Claude wherever you workClaude in Chrome; Claude for Microsoft 365§05
4 · Sharing and safetyWorking safely; validating skills for plugins; sharing with your team§08
5 · Wrap upCheck your understanding; quiz

Note that its module 2 language — "standing context", "encode your team's expertise" — is the same conceptual ladder as the maturity model in §12. Consistent vocabulary across Anthropic's material is useful; borrow it.

3 · AI Fluency: Framework and Foundations

Built with Prof. Joseph Feller (University College Cork) and Prof. Rick Dakan (Ringling College). 14 lectures. Its framework is the most reusable conceptual asset Anthropic publishes — see below.

The AI Fluency 4D framework

Framing line, verbatim: "Four interconnected competencies necessary to ensure our interactions with AI are effective, efficient, ethical and safe."

CompetencyDefinition (verbatim)
Delegation"Setting goals and deciding whether, when and how to engage with AI."
Description"Effectively describing goals to prompt useful AI behaviors and outputs."
Discernment"Accurately assessing the usefulness of AI outputs and behaviours."
Diligence"Taking responsibility for what we do with AI and how we do it."

Three modes of AI interaction

Three panels showing when the human's judgment is applied: Automation with the human only at the start, Augmentation with human and AI alternating throughout, and Agency with the human at configuration time and the AI acting repeatedly afterwards.
Fig 13.1 · Automation, Augmentation, Agency. Definitions verbatim. The arrangement by when judgment lands is the reading that connects this framework to the adoption ladder — see §14.
ModeDefinition (verbatim)Product analogue
Automation"AI executes specific tasks based on human instructions"Chat; a single-step prompt
Augmentation"Humans and AI collaborate as thinking partners"Interactive Cowork or Claude Code session
Agency"Humans configure AI to independently perform future tasks on their behalf"Skills, plugins, scheduled tasks, Routines, Dispatch
Why this is worth using

The three modes map almost perfectly onto the maturity ladder in §12 and onto the autonomy spectrum in §07. Automation ≈ Level 0. Augmentation ≈ Levels 1–2. Agency ≈ Levels 3–4. That gives you one vocabulary spanning literacy, product choice, and organisational maturity — which is exactly what a curriculum needs and what Anthropic never assembles in one place itself.

Also referred to as named modules: the Description-Discernment loop and the Delegation-Diligence loop.

Source dependency worth knowing

These two loop names come from the course module listing on Skilljar, which is behind a login. They do not appear in the public AI Fluency one-pager PDF, which presents the four competencies as a flat set with no loop structure. So any argument built on the loops — including the nested reading in §14 — rests on a source neither this pack nor an independent reviewer could open. Enrol in the free course and confirm before you teach the loop structure as Anthropic's.

The four AI Fluency competencies drawn as nested loops: an inner fast loop between Description and Discernment labelled per interaction, inside an outer slower loop between Delegation and Diligence labelled per decision.
Fig 13.2 · The 4Ds as nested loops. Competency definitions verbatim. The nesting is our reading, and it rests on loop names that appear only in the login-gated course — see the warning above.
Licence note

The framework artefacts are released under CC BY-NC-SA 4.0 (Copyright 2025 Rick Dakan, Joseph Feller, and Anthropic). The non-commercial clause restricts reuse of the diagram and wording as-is in paid material. Fine for internal and personal use; revisit if this ever becomes commercial output. Redrawing the concepts in your own visual language and mental model sidesteps the question entirely and produces better teaching material anyway.

Role guides and tutorials

Role guides (Help Centre, no signup)

Linked from /learn/claude-for-work: Claude for Engineering · Human Resources · Marketing · Product Management · Sales. Short, practical, and safe to hand out as-is.

Cowork tutorials — the role-specific walkthroughs

At claude.com/resources/tutorials, none linked from /learn:

The tutorials index also exposes Anthropic's own working taxonomy, which is a useful sanity-check on curriculum scope:

AxisValues
CategoryAI Fluency · Claude Science · Education · Engineering · Finance · Healthcare · HR · Life Sciences · Marketing · Nonprofits · Professional · Sales · Small Business
FeaturesArtifacts · Connectors · Projects · Research · Skills · Browser Use
ProductsClaude Science · Claude Security · Claude Design · Claude in PowerPoint · Claude in Excel · Claude Cowork · Claude in Chrome · Claude Code · Claude.ai

Certifications

Four role-based certifications, delivered through the Anthropic Partner Academy (training free), examined via Pearson Professional Assessments (proctored, identity-verified), with badges issued through Credly by Pearson.

CertificationTarget roleTests
Claude Certified Associate: FoundationsConsultants, project leads, professionals with business or technical expertise"practical, everyday use of Claude"
Claude Certified Developer: FoundationsEngineers building applicationsClaude API, tool use, agent development
Claude Certified Architect: FoundationsSolution architects designing agent systemsFoundation-level architecture
Claude Certified Architect: ProfessionalAdvanced architects at large enterprisesIntegration architecture, governance, evaluation

Structure, verbatim: "Every path to getting credentialed starts with a foundation-level certification and advances to the professional-level."

Scale figures Anthropic publishes: more than 36,000 consultants certified across more than 1,300 organizations; more than 400,000 people have completed Claude training through the Partner Academy this year; a $100M commitment to the Claude Partner Network for 2026.

Unverified

Certification pricing and syllabi could not be confirmed. claude.com/resources/certifications and claude.com/partners/certifications both return 404 — the first is a broken link on Anthropic's own courses page — and claude.com/certifications redirects to a login-gated page. Only the Associate track is aimed at non-engineers. Do not quote a price.

How Anthropic organises all this — and why it doesn't reconcile

Three overlapping taxonomies, which is worth knowing so you don't try to inherit a structure that isn't there:

  1. /learn persona split — Build with Claude · Claude for work · Claude for personal
  2. Course catalogue categories — Professional · Engineering · Education · Nonprofits
  3. Product-line tagging — AI Fluency · Claude.ai · Claude Cowork · Claude Code · Claude Platform · MCP

There is no numbered level system, no prerequisite chain, and no named executive persona anywhere in the public learning material. "101" is a naming convention, not a track. The only genuine maturity model Anthropic publishes is the five-level Cowork ladder in §12 — and it is not part of the Academy at all.

A telling detail: /learn/claude-for-work, the business page, lists five courses under "Claude for Work Courses" — two of which are for nonprofits and educators.

Gaps you will have to fill yourself

GapDetail
No executive personaCourses target developers, students, educators, nonprofits, small businesses and "builders." No C-suite course, no board-level module. The closest proxy merely mentions "leadership roles" in an SMB audience list.
No facilitation materialEverything is self-paced async video. No slide decks, facilitator guides, workshop scripts, discussion prompts or cohort structure. This pack's §11 exists precisely because Anthropic doesn't publish one.
No real assessmentMultiple-choice quizzes only. No capability rubric, no maturity self-assessment instrument, no pre/post diagnostic. The certifications are proctored but partner-gated and role-mismatched for internal audiences.
No ROI or business-case modelMetrics are listed; methods are not. No cost model, no TCO framework, no payback calculator, no benefits-realisation template. Customer figures are outcomes, not methodologies.
No Cowork communications kitClaude Code has a 20-message drip campaign and exec-sponsor templates. Cowork — the product executives will actually use — has nothing equivalent.
Nothing on organisational designNo guidance on headcount, job architecture, career ladders or performance management when Level 4 plugins absorb routine work. Anthropic's Economic Index work exists separately and is not integrated.
No leader-facing risk curriculumSecurity is documented for admins (RBAC, spend limits, telemetry). There is no executive module on AI risk appetite, model-failure liability, regulatory exposure or board reporting.
Shallow depth overallTotal Professional-track content is roughly five hours, nearly all sub-hour modules. These are on-ramps, not a curriculum.
What this means for your curriculum

Anthropic gives you good on-ramps (Claude 101, Intro to Cowork), a good conceptual frame (4D + three modes), and a good adoption model (§12). It gives you no depth, no facilitation, no assessment, no ROI method and no executive voice. Assign their courses as pre-work; the session itself is yours to build — and the gap list above is effectively a specification for what makes your material worth paying for.

Suggested pre-work assignment

#AssignTimeWhy
1AI capabilities and limitations15 minBaseline literacy; prevents the session being derailed by "does it hallucinate?"
2Claude 1011 hrShared vocabulary — projects, artifacts, skills, connectors — before you compare products
3Delegating your first task in Claude Cowork (video)5 minThey arrive having seen the interaction model, not just heard about it
4 (optional)AI fluency: Framework and foundations1.1 hrOnly if you intend to build the session on the 4D vocabulary

Total mandatory pre-work: ~80 minutes. All free, all self-paced, all certificated — the certificate is a surprisingly effective completion lever with senior audiences.