Skip to main content

Sources and Verification

Why this page exists

A learning plan is only reliable if its recommendations are real, current, and independently verifiable. This page documents every certification, framework, and key resource referenced in the plan, with links to primary sources (universities, Microsoft, standards bodies) so anyone — or a recruiter — can confirm them directly.

Verification method

  1. Primary sources only — official provider pages, not blogs or aggregators.
  2. Record the "verified on" date — certifications and URLs change; we note when each one was confirmed.
  3. Separate fact from recommendation — the existence and scope of an exam is a verifiable fact; sequence and timing are judgment.
  4. Re-verify before every exam — Microsoft regularly renews, renames, or retires credentials.

Last verified: 2026-07-24


Certifications (current, verified)

CertificationCodePrice (USD)Exam languagePrimary source
Azure FundamentalsAZ-900$99 (+ taxes)Spanish availablelearn.microsoft.com
Security, Compliance & Identity FundamentalsSC-900$99 (+ taxes)Spanish availablelearn.microsoft.com
Azure AI FundamentalsAI-900$99 (+ taxes)Spanish availablelearn.microsoft.com
GitHub FoundationsGH-900$99 (free prep)Englishlearn.microsoft.com
Azure AI Apps and Agents Developer AssociateAI-103verify on official pageVerifylearn.microsoft.com
AI-102 retired on June 30, 2026 — use AI-103

Microsoft retired the AI-102 exam on Jun 30, 2026. Its successor is AI-103 — Azure AI Apps and Agents Developer Associate (generative AI apps + multi-agent solutions with Microsoft Foundry, in Python). Official page: AI-103 · AI-103 study guide. Confirm status (beta/GA), price, and skills measured on the official page before scheduling.

About AI-103 for this profile

AI-103 (Associate level) is developer-oriented for people who build AI solutions with Python and Microsoft Foundry. For this profile (IT Manager targeting AI Program Manager / AI Governance / TPM) it is optional and advanced: it is worth it only if the target includes building AI solutions with code. For governance/program roles, vendor-neutral credentials (IAPP AIGP, PMI-CPMAI) + portfolio + frameworks (NIST/ISO) usually provide more return.


Vendor-neutral certifications (verified)

This plan is not "Microsoft first." These credentials are vendor-independent and often carry more weight in AI governance, risk, and security roles. Verify price and requirements at the source before paying.

CertIssuerPrerequisiteLanguagePrimary source (verified 2026-07)
AIGP — AI Governance ProfessionalIAPPNoneEnglishiapp.org/certify/aigp
PMI-CPMAI — Managing AIPMINoneSpanish (LATAM) + 9 morepmi.org
CEET — Certified Ethical Emerging TechnologistCertNexusNoneEnglishcertnexus.com
AAIA — Advanced in AI AuditISACA⚠️ Requires CISA (or another equivalent audit designation)Englishisaca.org/credentialing/aaia
Lead Auditor ISO/IEC 42001Various (accredited)NoneVariesiso.org/standard/81230
CC — Certified in CybersecurityISC2NoneSpanish availableisc2.org
Security+CompTIANoneSpanish availablecomptia.org
Verification of prerequisites and names
  • ISACA AAIA requires CISA (or another accepted audit/accounting designation) as a prerequisite — confirmed on the official page. Plan CISA first if you do not already have it.
  • PMI's official credential name is "PMI Certified Professional in Managing AI (PMI-CPMAI)" (previously referred to as AI Project Management).
  • We could not confirm an ISACA "AAISM" credential as a separate product; only AAIA is verified. If you see "AAISM," verify it directly at isaca.org before citing it.
  • Prices: vendor-neutral exams change price frequently; the ranges in the Plan Overview are directional (~AIGP $649–799, ~CPMAI $699 + membership, ~CEET $270–350, ~AAIA $459–599). Confirm at the official source.

Currency guardrail

Microsoft frequently updates skills measured, prices, and practice assessment availability. Before scheduling any exam, confirm on the official credential page: exam status, current syllabus, and whether there is a free practice assessment.


Governance and security frameworks (verified)

FrameworkIssuerOfficial identifierPrimary source
AI Risk Management FrameworkNIST (U.S.)AI RMF 1.0 (NIST AI 100-1)nist.gov
AI management systemISO/IECISO/IEC 42001:2023iso.org
EU AI ActEuropean UnionRegulation (EU) 2024/1689eur-lex.europa.eu
Zero Trust ArchitectureNISTSP 800-207csrc.nist.gov
Top 10 for LLMsOWASPOWASP Top 10 for LLM Appsowasp.org
Threats to AI systemsMITREMITRE ATLASatlas.mitre.org

Free and low-cost resources (trusted sources)

Verified options for building the same competencies at no cost or low cost — key for anyone in transition or career repositioning.

ResourceProviderCostCertificatePrimary source
Elements of AIUniversity of Helsinki + ReaktorFreeFreeelementsofai.com
Google AI EssentialsGoogleLow cost (free Coursera audit)Paidgrow.google/ai-essentials
AI for Everyone (Andrew Ng)DeepLearning.AI (Coursera)Free auditPaidcoursera.org
Generative AI for EveryoneDeepLearning.AI (Coursera)Free auditPaiddeeplearning.ai
AI learning pathsMicrosoft LearnFreeFree (badges/Applied Skills)learn.microsoft.com
Official prep + Practice AssessmentsMicrosoft LearnFreeN/Alearn.microsoft.com
Python para Todos (Py4E)Univ. of Michigan (Dr. Chuck)FreeOptionalpy4e.com
GitHub SkillsGitHubFreeN/Askills.github.com
Anthropic AcademyAnthropicFreeVariesanthropic.com/learn

AI Security — open and free tools (vendor-neutral)

Foundation of the Challenge Track: Applied AI Security. All are open source and runnable on your own machine.

Tool / ResourceWhat it doesCostPrimary source (verified 2026-07)
OWASP Top 10 for LLM AppsThe 10 security risks for LLMsFreegenai.owasp.org
MITRE ATLASAdversarial TTPs against AI/ML systemsFreeatlas.mitre.org
NIST AI RMF + GenAI Profile (600-1)AI risk management frameworkFreenist.gov · 600-1
garakLLM vulnerability scannerFree (Apache-2.0)github.com/NVIDIA/garak
PyRITGenerative AI red-teaming toolkitFree (MIT)github.com/Azure/PyRIT
PresidioPII detection/anonymization (DLP)Free (MIT)github.com/data-privacy-stack/presidio
OWASP Top 10:2025Web application security risks (SQLi, XSS, SSRF, auth, etc.)Freeowasp.org/Top10/2025
MITRE CWE Top 25Most dangerous software weaknesses (annually updated)Freecwe.mitre.org/top25
Semgrep OSSStatic code analysis (SAST) for AI-generated code reviewFree (LGPL 2.1)github.com/semgrep/semgrep
GitleaksSecret scanning (hardcoded credentials/API keys)Free (MIT)github.com/gitleaks/gitleaks
Location changes already verified (2026-07)

garak moved from leondz/garak to NVIDIA/garak. Presidio moved out of the Microsoft organization to data-privacy-stack/presidio (docs at data-privacy-stack.github.io/presidio). Always use the new URLs.

OWASP Top 10 edition change

OWASP released the Top 10:2025 edition, superseding the 2021 edition (owasp.org/Top10/ now redirects to it). Category numbers shifted — e.g., SSRF is folded into A01:2025 Broken Access Control rather than standing alone as it did in 2021. Always cite the current edition and re-check category numbers before an interview or exam, since OWASP periodically renumbers.

For someone between jobs

The free stack (Elements of AI + Microsoft Learn + Py4E + GitHub Skills + official practice assessments) covers almost the whole plan at no cost. The only mandatory spend is the certification exams (~$99 each) — and Microsoft periodically offers free vouchers through Virtual Training Days. Check the Microsoft Learn events page before paying.


Product changes already reflected

Learning plans go stale when vendors rename products. These are already corrected in the plan:

BeforeNowNote
Azure AI Studio (/azure/ai-studio/)Azure AI Foundry (/azure/ai-foundry/)Renamed by Microsoft; URLs updated
Power BI "Guided Learning"Modular Power BI pathsThe linear path was retired
Google AI Essentials (old URL)grow.google/ai-essentials/Current official URL
MeasureUp (paid practice tests)Official Microsoft Practice Assessments (free)Official, no-cost alternative

Learning design frameworks and market data

Methodology and Best Practices is based on these primary sources:

SourceWhat it contributesTypePrimary source
WEF Future of Jobs Report 2025In-demand skills and AI transformation data (86%, 39%, ~63%)Research reportweforum.org
70-20-10 modelExperience/social/formal balance in leader developmentFramework (CCL)ccl.org
Dunlosky et al. (2013)Effectiveness of active recall and spaced/retrieval practicePeer-reviewed studyjournals.sagepub.com
Google Career CertificatesDesign with projects + capstone + employer consortiumReference programgrow.google/certificates
AWS re/StartCohort design + scenario labsReference programaws.amazon.com
IBM SkillsBuildProject-based learning + stackable credentialsReference programskillsbuild.org

AI labor market data (2025–2026)

Use these in cover letters and interviews to support the transition with data. They are survey and job-posting aggregate data; treat them as directional, not guarantees.

Data pointSourceTypePrimary source
Job postings requiring AI skills pay an 28% salary premium ($18k/year)Lightcast (2025)Market reportlightcast.io
~Half of postings requiring AI skills are already outside IT (HR, finance)Indeed Hiring Lab (Jun 2025)Market reporthiringlab.org
Average total compensation in AI privacy/governance ~$200k; median AI governance specialist ~$152kIAPP (2025–26)Survey reportiapp.org
About WEF statistics

The figures (86% of employers expect AI transformation; 39% of core skills will change; ~63% cite the skills gap as the main barrier) come from the World Economic Forum's Future of Jobs Report 2025. They are employer survey data, not guaranteed projections. Google/AWS/IBM employment outcome percentages are reported by each provider and should be read as verifiable marketing, not independent studies.

About the 70-20-10 model

The 70/20/10 proportions are a guiding philosophy based on research by the Center for Creative Leadership (McCall, Lombardo, and Eichinger, 1988), not a statistically validated formula. Use them as a balance guide, not an exact rule.


Microsoft Copilot Studio (verified)

Official resources and credentials for the priority Copilot Studio module in Phase 1. Copilot Studio evolves quickly (autonomous agents, generative actions); re-verify before scheduling a credential.

Resource / CredentialTypeCostPrimary source
Official Copilot Studio documentationDocsFreelearn.microsoft.com
Learning Path: Create agents in Microsoft Copilot Studio (9 modules)Learning Path with labsFreelearn.microsoft.com
Copilot Studio Agent AcademyPractical curriculumFreemicrosoft.github.io/agent-academy
Agent in a Day — WorkshopGuided workshopFreelearn.microsoft.com
Applied Skill: Create agents in Microsoft Copilot StudioCredential (interactive lab)Free / low costlearn.microsoft.com
Applied Skill: Enhance agents with autonomous capabilitiesCredential (interactive lab)Free / low costlearn.microsoft.com
Copilot Studio SamplesSample codeFreegithub.com/microsoft/CopilotStudioSamples
About Applied Skills

They are interactive lab-based credentials (not multiple-choice exams): you demonstrate the skill by performing real tasks in a sandbox environment. Both official pages were confirmed and updated by Microsoft in 2025-2026. Copilot Studio is the evolution of Power Virtual Agents.


Note about salary ranges

The salary ranges in the Plan Overview are directional (U.S. market, 2025-2026), not guarantees. They vary strongly by country, industry, company size, and seniority. For role/region-specific figures, consult live data sources such as:

  • Levels.fyi — tech compensation by level and company
  • Glassdoor — ranges by role and location
  • Robert Half / Michael Page market reports (verify the year)

Honesty log

  • Verified facts: existence, code, and language of Microsoft certifications (~$99 fundamentals); AI-102 retired on Jun 30, 2026, successor AI-103; existence and prerequisites of vendor-neutral credentials (IAPP AIGP, PMI-CPMAI — exam in LATAM Spanish —, CertNexus CEET, ISACA AAIA requires CISA); framework identifiers (NIST AI RMF 1.0, ISO/IEC 42001:2023, EU AI Act = Regulation (EU) 2024/1689, NIST SP 800-207); current locations of open tools (garak → NVIDIA/garak; Presidio → data-privacy-stack/presidio).
  • Not verified / marked: we could not confirm an ISACA "AAISM" credential distinct from AAIA; paid exam prices change and are directional.
  • Judgment (not fact): the phase sequence, weekly timing, and target role selection are recommendations.
  • Design frameworks: the 70-20-10 model (CCL) and WEF Future of Jobs 2025 data support the methodology; the 70/20/10 proportions are a guiding philosophy, not a validated formula, and WEF statistics are survey data with year/source cited.
  • Directional pending verification: salary ranges and any market statistics.
  • Prices and URLs change: always confirm at the source before enrolling or scheduling an exam.