open/ms-ai-architect
1
0
Fork 0
Code Issues Pull requests Releases Activity

feat(ms-ai-architect): R7.4 bølge 4 — payload-30..39 dømt+ingestet (190 claims, 10 flagged/0 pass, 55 flagg; ledger 172→182) [skip-docs]

Browse source 
Avslutter responsible-ai/ (slot 30-33) og starter ms-ai-infrastructure/bcdr/
(slot 34-39). Ingen pass-record, så re-judge-gaten utløses ikke og
apply-verified-stamp.mjs er ikke kjørt (verified uendret på 12).

validate-result.mjs: falsk positiv rettet — «opplyser» sto i
folding-heuristikkens ordliste, men er ordinært norsk uten foldet vokal
(ae/oe/aa) og blokkerte et korrekt result-32. Oppføringen er fjernet med
begrunnelse i koden. Regresjon: alle 39 resultater validerer, og
kontrolltest bekrefter at ekte folding (paa/staar/noeyaktig/loesning/
maaned/ogsaa/naar/aapen/stoette/tjenesteomraade) fortsatt fanges.

Suite 947/947 grønn. R7.4: 39/49 dømt. Flagg totalt 657 → 712.
This commit is contained in:
Kjell Tore Guttormsen 2026-07-31 20:57:05 +02:00
commit 533d451cd2
1 changed files with 771 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

772
scripts/kb-eval/data/judge-pass-manifest.json
View file

@ -4,7 +4,7 @@
"derived_from": "scripts/kb-update/data/full-pass-worklist.json (243 due, all never-verified)",
"cadence": "R7–R10 (5 økter × ~49, 8–10 samtidige)",
"batch": "R7.1",
"count": 172,
"count": 182,
"generated": "2026-07-18"
},
"files": [
@ -9311,6 +9311,776 @@
}
]
},
{
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-policy-development.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 19,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-policy-development.md#3",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/azure/machine-learning/concept-responsible-ai",
"evidence_quote": "",
"reason": "Ingen learn-side sier at Privacy & Security-domenet i RAI-standarden krever differential privacy, data minimization og secure model deployment; kravene ligger i RAI-standardens PDF utenfor learn, og Azure ML omtaler differential privacy som et valgfritt open source-verktøy (SmartNoise), ikke som et krav.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-policy-development.md",
"line": 50,
"claim": "Privacy & Security-domenet i Microsoft Responsible AI Standard krever at team implementerer differential privacy | data minimization | secure model deployment.",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-policy-development.md#9",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/azure/cloud-adoption-framework/scenarios/ai/platform/governance",
"evidence_quote": "When you deploy an Azure landing zone, select the policy initiative under the Workload Specific Compliance category. Available options include: Azure Machine Learning, Azure AI Search, Azure AI Bot Service",
"reason": "Den kanoniske siden som teller opp AI-policy-initiativene lister kun tre (ML, AI Search/Cognitive Services, Bot Service); det påståtte Azure OpenAI Guardrails-initiativet er fraværende.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-policy-development.md",
"line": 223,
"claim": "Azure Policy-initiativer for AI omfatter: Azure OpenAI Guardrails-initiativ | Azure Machine Learning ML guardrails | Azure AI Search Cognitive Services guardrails | Azure AI Bot Service Bot guardrails.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-policy-development.md#15",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/microsoft-copilot-studio/guidance/sec-gov-phase5",
"evidence_quote": "Capacity and quota monitoring: Monitor and manage your agents' message usage, capacity, and other solution dependencies (for example, Power Automate, prompt builder) in the Power Platform admin center to prevent overages and service throttling.",
"reason": "Dataresidens, ISO/SOC/HIPAA-sertifisering og dokumenterte best practices holder, men analytics-dashbordet teller opp aktive brukere, sesjoner, utfall, tilfredshet og temaer — Copilot Studio måler forbruk i meldinger/Copilot Credits, ikke token-bruk, og ingen side nevner identifisering av høykost-skills.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-policy-development.md",
"line": 269,
"claim": "Copilot Studios governance-funksjoner: data location-kontroller som respekterer datasuverenitetskrav | compliance-sertifiseringer ISO, SOC, HIPAA | analytics dashboard som overvåker token-bruk og identifiserer høykost-skills | dokumenterte best practices for sikkerhet og governance.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-policy-development.md#17",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/purview/purview-billing-models",
"evidence_quote": "",
"reason": "Ingen learn-side stiller opp dette settet av lisensmodeller; enkeltdeler er støttet (Purview per-user + pay-as-you-go per GB, Azure Monitor per GB ingested pluss retention), mens prisstruktur for Azure Policy, Foundry, Copilot Studio og Defender kun finnes på JS-rendrede azure.microsoft.com-prissider — ingenting motsier påstanden.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-policy-development.md",
"line": 417,
"claim": "Lisensmodeller for Microsoft governance-verktøy: Azure Policy inkludert i Azure subscription | Microsoft Purview per GB data + per bruker | Microsoft Foundry pay-as-you-go (compute, storage, API-kall) | Copilot Studio per bruker/måned eller per sesjon | Azure Monitor per GB ingested + retention | Microsoft Defender for Cloud per ressurs.",
"disposition": "unsourced"
}
]
},
{
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-training-awareness.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 17,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-training-awareness.md#1",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/azure/cloud-adoption-framework/scenarios/ai/plan",
"evidence_quote": "Use the AI learning hub platform for free AI training, certifications, and product guidance. Set certification goals such as Azure AI Fundamentals, and Azure Data Scientist Associate certifications.",
"reason": "Den siterte Acquire AI skills-seksjonen lister interne læringsprogrammer, rekruttering og partnerskap, men definerer ingen trelags-taksonomi AI Awareness / AI Literacy / AI Competency; søk på Learn finner heller ingen slik Microsoft-modell, så fraværet i den kanoniske siden teller som motbevis.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-training-awareness.md",
"line": 42,
"claim": "Microsoft definerer et trelags opplæringsopplegg for Responsible AI: AI Awareness (alle ansatte) | AI Literacy (knowledge workers, produkteiere) | AI Competency (utviklere, data scientists, arkitekter).",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-training-awareness.md#2",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/azure/cloud-adoption-framework/scenarios/ai/plan",
"evidence_quote": "Set certification goals such as Azure AI Fundamentals, and Azure Data Scientist Associate certifications.",
"reason": "Siden nevner kun Azure AI Fundamentals og Azure Data Scientist Associate som sertifiseringsmål — verken AI-102 eller et AI Competency-nivå står der — og Microsofts retirement-register fører nå AI-900 (30. juni 2026), AI-102 (30. juni 2026) og DP-100 (1. juni 2026) som pensjonerte eksamener.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-training-awareness.md",
"line": 48,
"claim": "Nivået AI Competency (utviklere, data scientists, arkitekter) dekkes av sertifiseringskursene AI-900 | AI-102 | DP-100.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-training-awareness.md#3",
"judge_verdict": "not_grounded",
"rule": "R3",
"evidence_url": "https://learn.microsoft.com/training/modules/embrace-responsible-ai-principles-practices/",
"evidence_quote": "Embrace responsible AI principles and practices - Module - 9 Units",
"reason": "Alle tre er merket Module på Microsoft Learn, ikke Learning Path; learning path er en egen ressurstype som bunter moduler (f.eks. Responsible use of artificial intelligence in education - Learning Path - 3 Modules), så påstandens organiserende enhet er feil.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-training-awareness.md",
"line": 50,
"claim": "Microsoft Learn tilbyr tre strukturerte learning paths for Responsible AI: Embrace Responsible AI Principles and Practices | Apply Responsible AI Principles in Learning Environments | Implement a Responsible Generative AI Solution in Microsoft Foundry.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-training-awareness.md#6",
"judge_verdict": "not_grounded",
"rule": "R4",
"evidence_url": "https://learn.microsoft.com/credentials/certifications/azure-ai-fundamentals/",
"evidence_quote": "Required exams: AI-901",
"reason": "Sertifiseringssiden oppgir nå AI-901 som påkrevd eksamen, og Microsofts eksamens-retirement-register fører AI-900 som pensjonert 30. juni 2026 (studieguiden: This exam was retired on June 30, 2026) — påstanden om at AI-900 ikke har noen formell utløpsdato matcher bare en utgått rad.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-training-awareness.md",
"line": 77,
"claim": "AI-900 (Azure AI Fundamentals) har ingen formell utløpsdato.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-training-awareness.md#8",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/credentials/certifications/azure-data-scientist/",
"evidence_quote": "This certification and the renewal assessment are retired.",
"reason": "Den bærende delen gjeldende sertifisering motsies direkte av advarselen på sertifiseringssiden, og retirement-registeret fører DP-100 som pensjonert 1. juni 2026 — at CAF-siden fortsatt nevner den som mål endrer ikke statusen.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-training-awareness.md",
"line": 175,
"claim": "Azure Data Scientist Associate er en gjeldende Microsoft-sertifisering (credentials/certifications/azure-data-scientist) som anbefales som minimumsbaseline for data scientists.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/responsible-ai-training-awareness.md#15",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/viva/learning/set-up-viva-learning#prerequisites",
"evidence_quote": "Your Microsoft 365 or Office 365 Enterprise subscription includes Viva Learning with basic features.",
"reason": "Lisensgulvet påstanden setter er for høyt: Viva-tjenestebeskrivelsens tilgjengelighetstabell gir Yes for basisfunksjonene også i Office 365 E1, Microsoft 365 F1/F3 og Microsoft 365 Business Basic og Standard, så basis-Viva Learning krever ikke E3/E5 eller Business Premium.",
"file": "skills/ms-ai-governance/references/responsible-ai/responsible-ai-training-awareness.md",
"line": 381,
"claim": "Viva Learning (basic) krever Microsoft 365 E3/E5 eller Business Premium og er inkludert i disse lisensene.",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 22,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#2",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/how-to-machine-learning-interpretability?view=azureml-api-2",
"evidence_quote": "It provides multiple views into a model's behavior: Global explanations: For example, what features affect the overall behavior of a loan allocation model? Local explanations: For example, why was a customer's loan application approved or rejected? You can also observe model explanations for a selected cohort as a subgroup of data points.",
"reason": "De tre nivåene stemmer, men verktøy-tilordningen er feil: siden legger BÅDE lokale og kohort-forklaringer til selve interpretability-komponenten, mens counterfactual what-if er en separat DiCE-basert komponent for feature-perturbasjoner — to bærende deler er feilkoblet.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 91,
"claim": "Model Interpretability gir tre nivåer av forklaring med tilhørende Microsoft-verktøy: Global explanations (Azure ML Interpretability component) | Local explanations (Counterfactual What-If) | Cohort explanations (Responsible AI Dashboard)",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#4",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/ai-agents/responsible-ai-across-organization",
"evidence_quote": "Empower a cross-functional governance team. Establish an AI Center of Excellence or ethics committee that includes representatives from legal, security, product, and engineering teams. This group defines standards and provides consultative support rather than acting as gatekeepers. Give this team executive sponsorship and clear authority to enforce policies when necessary.",
"reason": "Den kanoniske siden beskriver ÉN enhet — AI Center of Excellence = cross-functional governance team, med executive sponsorship — ikke et AI Governance Board over tre underenheter; verken AI Governance Board eller Incident Response Team finnes som navngitte enheter (incident response er en prosedyre, ikke et organ).",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 170,
"claim": "Microsofts governance-struktur for AI i organisasjoner består av et AI Governance Board (executive sponsorship) over tre enheter: AI Center of Excellence | Cross-Functional Governance Team | Incident Response Team",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#6",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/concept-responsible-ai?view=azureml-api-2",
"evidence_quote": "Data-driven insights, which help stakeholders understand causal treatment effects on outcomes using historical data only... These insights come from the causal inference component of the Responsible AI dashboard. Model-driven insights, which answer user questions (such as What can I do to get a different outcome from your AI next time?) so they can take action. These insights are provided through the counterfactual what-if component of the Responsible AI dashboard.",
"reason": "Alle seks verktøyene finnes, men den bærende glossen er feil: counterfactuals hører til den separate counterfactual what-if-komponenten (DiCE), mens causal inference (EconML) handler om kausale behandlingseffekter — claimet slår dem sammen og utelater counterfactual what-if som eget verktøy.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 374,
"claim": "Azure Machine Learning tilbyr disse Responsible AI-verktøyene: Responsible AI Dashboard (end-to-end model assessment) | Responsible AI Scorecard (PDF-eksport av dashboard-innsikt) | Model Interpretability | Fairness Assessment (group-based performance metrics) | Error Analysis (cohort-based error distribution) | Causal Inference (what-if-analyse for counterfactuals)",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#9",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/ai-agents/governance-security-across-organization",
"evidence_quote": "Require every AI agent to be recorded in a single organizational inventory. Track ownership, purpose, platform, and access scope. Treat agents as managed organizational resources. Agent 365 provides an Agent Registry when adopted... Use Microsoft Entra Agent ID to assign identity, permissions, and lifecycle controls.",
"reason": "Siden tilskriver unik agent-identitet til Microsoft Entra Agent ID og registeret til Agent 365 — ikke til Copilot Studio — og sporer ownership/purpose/platform/access scope, ikke owner/version/lifecycle status; Copilot Studios egen fasilitering er Application Insights-telemetri og Copilot Credits-analytics, ikke Log Analytics-hendelseslogg med token/compute per agent.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 437,
"claim": "Copilot Studio har disse stakeholder-kommunikasjonsfunksjonene: agent observability med unik identitet per agent (owner, version, lifecycle status) | centralized logging av nøkkelhendelser til Azure Log Analytics | cost tracking av token consumption og compute usage per agent | user disclosure der agenter identifiserer seg som AI og ikke menneske",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#11",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/power-platform/guidance/coe/governance-components",
"evidence_quote": "Compliance detail request emails are sent for apps and chatbots. This flow sends an email to users who have apps in the tenant that aren't compliant with the following thresholds...",
"reason": "Den kanoniske opplistingen dekker apper, flows, chatbots, custom connectors og miljøer — AI Builder-modeller er fraværende fra inventaret — og compliance-varslene går til eierne/makerne, ikke til et governance-team ved high-risk deployments.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 471,
"claim": "Power Platform CoE Starter Kit gir governance-kapabiliteter: inventory av alle AI Builder-modeller i tenant | compliance checks (er modellen i prod uten review?) | automatiske varsler til governance team ved high-risk deployments",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#12",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/concept-responsible-ai-scorecard?view=azureml-api-2",
"evidence_quote": "",
"reason": "Scorecard-siden beskriver funksjonen og preview-status, men sier ingenting om kostnad eller at PDF-eksporten er gratis; ingen Learn-side oppgir denne prisen.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 577,
"claim": "Responsible AI Scorecard er inkludert i Azure ML Workspace uten ekstra kostnad, og PDF-eksporten genereres gratis fra dashboard-data",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#13",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/azure/machine-learning/concept-responsible-ai",
"evidence_quote": "",
"reason": "Claimet handler om hvilken modell KB-filens eget kostnadseksempel bruker — en egenskap ved dokumentet, ikke ved Microsoft-produktet; ingen Learn-side kan bekrefte eller motsi det.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 604,
"claim": "Azure OpenAI-modellen som brukes i kostnadseksempelet for chatbot er GPT-4",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#15",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/office365/servicedescriptions/microsoft-agent-365/microsoft-agent-365",
"evidence_quote": "Microsoft Agent 365 is available as a stand-alone subscription for use with eligible Microsoft 365 subscriptions and is also included with Microsoft 365 E7.",
"reason": "Agent 365 lisensieres per bruker som egen subscription eller via Microsoft 365 E7 (med E5 som forutsetning for nye kjøp fra 2026-06-01); Copilot Studio finnes ikke i opplistingen over lisenser som inkluderer Agent 365.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 617,
"claim": "Agent observability (Microsoft Agent 365) er inkludert i Copilot Studio-lisensen",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/stakeholder-communication-ai-decisions.md#21",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/concept-responsible-ai-scorecard?view=azureml-api-2",
"evidence_quote": "This feature is currently in public preview. This preview version is provided without a service-level agreement, and we don't recommend it for production workloads. Certain features might not be supported or might have constrained capabilities.",
"reason": "Hele scorecardet er i public preview — også tittelen bærer (preview) — så den bærende delen status GA med public preview kun for enkelte funksjoner er direkte motsagt.",
"file": "skills/ms-ai-governance/references/responsible-ai/stakeholder-communication-ai-decisions.md",
"line": 797,
"claim": "Responsible AI Scorecard er dokumentert som preview — status GA med public preview for enkelte funksjoner (verifisert 2026-02)",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 22,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#3",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/azure/foundry/concepts/foundry-models-overview",
"evidence_quote": "Some of the details available in the model card are: Quick facts: Key information about the model at a quick glance",
"reason": "Kun første del holder (model catalog har innebygde model cards); den kanoniske model card-oppregningen nevner verken automatisk synkronisering av Hugging Face model cards eller noen mal for å generere egne model cards for custom modeller — to angitte, bærende deler mangler dekning.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 82,
"claim": "Microsofts model card-implementasjon omfatter Microsoft Foundry model catalog med innebygde model cards for pretrained modeller | Hugging Face-integrasjon der model cards synkroniseres automatisk | mal for å generere egne model cards for custom modeller.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#4",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/how-to-responsible-ai-scorecard",
"evidence_quote": "The Responsible AI scorecard is a PDF summary of key insights from your Responsible AI dashboard. The first summary segment of the scorecard gives you an overview of the machine learning model and the key target values you set",
"reason": "Den kanoniske gjennomgangen av scorecard-segmentene lister model overview, data analysis, model performance, cohorts, top important factors, fairness insights og causal insights — verken Error analysis eller Counterfactual analysis er scorecard-komponenter, så to angitte bærende deler er feil.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 114,
"claim": "Responsible AI Scorecard inneholder komponentene Model overview | Fairness assessment | Model interpretability | Error analysis | Counterfactual analysis | Causal inference | Data quality.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#8",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/en-us/azure/foundry/responsible-ai/openai/transparency-note",
"evidence_quote": "Model group | Text / code | Vision | Audio / Speech ... GPT-image-1 series ... GPT-image-2 ... Transcribe ... GPT-4 Turbo with Vision ... o1 series ... o3/o3-pro",
"reason": "Jeg hentet den kanoniske siden som ville oppregnet dem: Azure OpenAI har ÉN samlet transparency note med faner for tekst/vision/audio, ikke egne notiser per modell — og modelltabellen lister verken DALL-E 3 eller Whisper (erstattet av GPT-image-serien og Transcribe).",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 386,
"claim": "Azure OpenAI Service har egne transparency notes per modell, blant annet GPT-4 | DALL-E 3 | Whisper | o1.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#11",
"judge_verdict": "not_grounded",
"rule": "R1",
"evidence_url": "https://learn.microsoft.com/azure/foundry/concepts/foundry-models-overview",
"evidence_quote": "The catalog includes over 10,000 models, with approximately 50 new models published each month.",
"reason": "Nedre grense over 1500 overskrides grovt av den faktiske verdien over 10 000 (~7x) — bounden er beslutningsendrende misvisende etter lower-bound-policyen.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 407,
"claim": "Microsoft Foundry model catalog inneholder over 1500 pretrained modeller med model cards.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#12",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/microsoft-copilot-studio/system-service-card-copilot-studio",
"evidence_quote": "Human Oversight Guidance: Users are advised to review AI-generated outputs and automated actions before applying them in high-stakes scenarios. This practice mitigates risks of overreliance and ensures accountability.",
"reason": "Citations for generative answers og lenke til personvernerklæring er dokumentert, men den kanoniske oppregningen av innebygde sikkerhetskomponenter i Copilot Studio nevner verken et «Powered by AI»-merke i chat-vinduet eller bekreftelsesdialoger før sensitive plugin-handlinger — to angitte bærende deler mangler.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 426,
"claim": "Microsoft Copilot Studio har innebygde disclosures: «Powered by AI»-merke i chat-vinduet | attribusjonslenker til kildedokumenter for generative answers | bekreftelsesdialoger før sensitive plugin-handlinger | lenke til personvernerklæring i bot-innstillingene.",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#14",
"judge_verdict": "not_grounded",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/foundry/responsible-ai/openai/transparency-note",
"evidence_quote": "Non-English translations are provided for convenience only. Please consult the EN-US version of this document for the definitive version.",
"reason": "Siden slår fast at Microsoft selv leverer ikke-engelske oversettelser (engelsk er kun den autoritative versjonen), noe som motsier at notisene er tilgjengelige kun på engelsk.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 497,
"claim": "Microsofts Transparency Notes er tilgjengelige kun på engelsk (kan oversettes av kunden).",
"disposition": "outdated"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#15",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/concept-responsible-ai-scorecard",
"evidence_quote": "",
"reason": "Verken konseptsiden eller how-to-siden for scorecard omtaler språkstøtte i det hele tatt, så påstanden om manglende norsk kan verken bekreftes eller avkreftes.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 498,
"claim": "Responsible AI Scorecard støtter ikke norsk i preview — manuell oversettelse er nødvendig.",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#16",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/concept-responsible-ai-dashboard",
"evidence_quote": "",
"reason": "Dashboard-siden beskriver komponenter og begrensninger, men sier ingenting om kostnad eller lisenskrav; ingen Learn-side oppgir at dashboardet er inkludert i compute-kostnaden.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 523,
"claim": "Responsible AI dashboard er inkludert i Azure ML compute-kostnaden og krever ingen ekstra lisens.",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#17",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/concept-responsible-ai-scorecard",
"evidence_quote": "",
"reason": "Preview-merknaden sier kun at funksjonen leveres uten SLA og ikke anbefales for produksjon — ingenting om at den er gratis; prisinformasjon er ikke oppgitt på Learn.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 528,
"claim": "Responsible AI Scorecard er gratis som preview-funksjon.",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#18",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/machine-learning/concept-responsible-ai-scorecard",
"evidence_quote": "",
"reason": "Siden omtaler arkivering i Azure Machine Learning Run History, men ingen Learn-side oppgir et 90-dagers gratisvindu eller påfølgende standard storage-prising.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 530,
"claim": "Scorecards lagret i Azure ML Run History er gratis i 90 dager, deretter gjelder standard storage-prising.",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#19",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/foundry/responsible-ai/openai/transparency-note",
"evidence_quote": "",
"reason": "Verken transparency-notisen eller content filter-sidene oppgir prising for Azure OpenAI-annotasjoner; kostnadsspørsmålet ligger på JS-rendrede prissider som ikke kan hentes.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 538,
"claim": "Content Safety-annotasjoner er inkludert i Azure OpenAI API-prisingen uten ekstra kostnad.",
"disposition": "unsourced"
},
{
"id": "ms-ai-governance/responsible-ai/transparency-documentation-standards.md#20",
"judge_verdict": "not_grounded",
"rule": "R4",
"evidence_url": "https://learn.microsoft.com/en-us/azure/foundry/responsible-ai/openai/transparency-note",
"evidence_quote": "see the OpenAI o1 System Card, o3-mini System Card, o3/o4-mini System Card, Deep Research System Card, and GPT-5 System Card.",
"reason": "Den gjeldende versjonen av notisen er oppdatert langt forbi o3/o4-mini og Deep Research — den siterer GPT-5 System Card og dekker GPT-5-serien, GPT-5.1-Codex-Max, GPT-image-2 og GPT-Realtime-2, så «sist oppdatert med» treffer en foreldet tilstand.",
"file": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md",
"line": 717,
"claim": "Transparency note for Azure OpenAI er sist oppdatert med o3/o4-mini og Deep Research system cards.",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-infrastructure/references/bcdr/ai-foundry-disaster-recovery-planning.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 25,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-infrastructure/bcdr/ai-foundry-disaster-recovery-planning.md#16",
"judge_verdict": "not_grounded",
"rule": "R7",
"evidence_url": "https://learn.microsoft.com/azure/ai-foundry/openai/how-to/deployment-types",
"evidence_quote": "**SKU name in code:** `DataZoneStandard`",
"reason": "Data Zone Standard er en reell deployeringstype for gpt-4o, men den dokumenterte kodestrengen er `DataZoneStandard` — ingen Learn-side bruker `data_zone_standard`, og strengen er selve påstanden (R7 load-bearing-unntaket).",
"file": "skills/ms-ai-infrastructure/references/bcdr/ai-foundry-disaster-recovery-planning.md",
"line": 146,
"claim": "`data_zone_standard` er en gyldig deployment-type (deployment_type) for en gpt-4o-deployering.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/ai-foundry-disaster-recovery-planning.md#18",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/azure/ai-foundry/openai/quotas-limits",
"evidence_quote": "",
"reason": "Ingen Learn-side oppgir 120 000 TPM for denne deployeringen — kvoteallokeringen er en kundekonfigurert verdi, og kvotesidene angir bruks-/standardgrenser, ikke denne verdien.",
"file": "skills/ms-ai-infrastructure/references/bcdr/ai-foundry-disaster-recovery-planning.md",
"line": 151,
"claim": "Kvoten for gpt-4o-deployeringen settes til 120 000 TPM i primærregion og 120 000 TPM i sekundærregion.",
"disposition": "unsourced"
},
{
"id": "ms-ai-infrastructure/bcdr/ai-foundry-disaster-recovery-planning.md#20",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/azure/ai-foundry/openai/concepts/models#fine-tuning-models",
"evidence_quote": "| Model ID | Standard Training Regions | Global Training (Preview) | ... | `gpt-4o` (2024-08-06) | East US2 North Central US Sweden Central | - |",
"reason": "Sweden Central er en standard (regional) treningsregion, men Norway East finnes ikke i noen «Standard Training Regions»-kolonne — Norway East står kun i listen over Global training-regioner, så én bærende del er feil.",
"file": "skills/ms-ai-infrastructure/references/bcdr/ai-foundry-disaster-recovery-planning.md",
"line": 175,
"claim": "Regional training for fine-tuning er tilgjengelig i Norway East og Sweden Central.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/ai-foundry-disaster-recovery-planning.md#24",
"judge_verdict": "not_grounded",
"rule": "R3",
"evidence_url": "https://learn.microsoft.com/azure/foundry/how-to/create-resource-template",
"evidence_quote": "This template creates the following resources: microsoft.cognitiveservices/accounts, microsoft.cognitiveservices/accounts/projects",
"reason": "API-versjonen 2024-04-01 finnes for Microsoft.MachineLearningServices/workspaces, men ressursmodellen er byttet ut: et Foundry-prosjekt deployeres i dag som Microsoft.CognitiveServices/accounts/projects, mens workspaces-typen tilhører den avviklede hub-modellen.",
"file": "skills/ms-ai-infrastructure/references/bcdr/ai-foundry-disaster-recovery-planning.md",
"line": 236,
"claim": "AI Foundry project deployeres med resource-typen `Microsoft.MachineLearningServices/workspaces` og API-versjon 2024-04-01.",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 17,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-infrastructure/bcdr/backup-recovery-strategies-ai-workloads.md#2",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/en-us/azure/backup/backup-overview",
"evidence_quote": "SQL Server in Azure VMs - Back up SQL Server databases running on Azure VMs",
"reason": "Den kanoniske «What can I back up?»-opplistingen dekker VM-er, Managed Disks, Azure Files, Azure Blobs, PostgreSQL og SQL Server i Azure VM-er, men Azure SQL Database (PaaS) står ikke i listen — den har egne automatiserte backups, så «Azure SQL Database» som Azure Backup-ressurs er feil.",
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"line": 61,
"claim": "Azure Backup støtter følgende ressurser som er relevante for AI-arbeidsbelastninger: Azure Virtual Machines (GPU/compute) | Azure Managed Disks | Azure Files (SMB/NFS) | Azure Blob Storage (operational backup) | Azure SQL Database | Azure Database for PostgreSQL.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/backup-recovery-strategies-ai-workloads.md#4",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/security/benchmark/azure/baselines/azure-ai-foundry-security-baseline",
"evidence_quote": "Azure backup can be configured on the storage account attached to the AI project.",
"reason": "Sikkerhetsbaselinens BR-1-tabell merker «Azure Backup» som Supported = True for Foundry (kun «Service Native Backup Capability» er False), så den lastbærende delen «verken støtte i Azure Backup» er motsagt selv om AI Search-halvdelen stemmer.",
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"line": 74,
"claim": "Microsoft Foundry og Azure AI Search har verken støtte i Azure Backup eller nativ backup-mekanisme.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/backup-recovery-strategies-ai-workloads.md#6",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/en-us/azure/cosmos-db/continuous-backup-restore-introduction",
"evidence_quote": "The time window available for restore (also known as the retention period) depends on the continuous backup tier you select: 35 days, 30 days, or 7 days. The `Continuous35Days` tier is currently in preview.",
"reason": "Den kanoniske siden lister tre nivåer (7, 30 og 35 dager) og gjentar dem under «Compare the continuous backup tiers», mens påstanden hevder et lukket sett på to — nivålisten er utvidet siden påstanden ble skrevet.",
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"line": 151,
"claim": "Azure Cosmos DB continuous backup tilbys i to nivåer: Continuous 7-day med 7 dagers retensjonsperiode | Continuous 30-day med 30 dagers retensjonsperiode.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/backup-recovery-strategies-ai-workloads.md#7",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/cosmos-db/continuous-backup-restore-introduction",
"evidence_quote": "Continuous backup mode supports two ways to restore deleted containers and databases. They can be restored into a new account or into an existing account.",
"reason": "Listen over konfigurasjoner som ikke gjenopprettes (brannmur, VNet, RBAC, private endpoints, lagrede prosedyrer, triggere, UDF-er) stemmer, men den lastbærende delen «det opprettes alltid en ny konto» er motsagt av gjeldende avsnitt om gjenoppretting til en eksisterende konto.",
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"line": 169,
"claim": "Ved point-in-time-gjenoppretting av Azure Cosmos DB opprettes det alltid en ny konto, og følgende konfigurasjoner gjenopprettes ikke automatisk og må rekonfigureres: brannmurregler | VNet-innstillinger | RBAC-tildelinger | private endpoints | lagrede prosedyrer | triggere | UDF-er.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/backup-recovery-strategies-ai-workloads.md#12",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/storage/blobs/versioning-overview",
"evidence_quote": "Having a large number of versions per blob can increase the latency for blob listing operations. Microsoft recommends maintaining fewer than 1000 versions per blob.",
"reason": "Ingen Learn-side sier «ubegrenset», og den lastbærende parentesen «kun begrenset av lagringskostnad» er motsagt: Microsoft anbefaler under 1000 versjoner per blob av ytelseshensyn, ikke kostnadshensyn.",
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"line": 200,
"claim": "Azure Blob-versjonering gir ubegrenset antall blob-versjoner (kun begrenset av lagringskostnad).",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/backup-recovery-strategies-ai-workloads.md#16",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/reliability/regions-list",
"evidence_quote": "",
"reason": "Learn beskriver Norway West som en tilgangsbegrenset region reservert for Norway East-kunder, men ingen hentet side sier «begrenset tjenestestøtte», og ingen side anbefaler Sweden Central som alternativ sekundærregion for Norge — påstanden kan verken bekreftes eller motbevises.",
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"line": 369,
"claim": "Norway West har begrenset tjenestestøtte, og Sweden Central anbefales som alternativ sekundærregion.",
"disposition": "unsourced"
},
{
"id": "ms-ai-infrastructure/bcdr/backup-recovery-strategies-ai-workloads.md#17",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/azure/backup/policy-reference",
"evidence_quote": "Azure Backup should be enabled for Virtual Machines | Ensure protection of your Azure Virtual Machines by enabling Azure Backup. Azure Backup is a secure and cost effective data protection solution for Azure. | AuditIfNotExists, Disabled",
"reason": "GUID-en 013e242c-8828-4970-87b3-ab247555486d er «Azure Backup should be enabled for Virtual Machines» med effektene AuditIfNotExists/Disabled — en ren revisjonspolicy; det er andre definisjoner (83644c87-… / 98d0b9f8-…) som faktisk konfigurerer backup til en Recovery Services vault.",
"file": "skills/ms-ai-infrastructure/references/bcdr/backup-recovery-strategies-ai-workloads.md",
"line": 380,
"claim": "Den innebygde Azure Policy-definisjonen med ID 013e242c-8828-4970-87b3-ab247555486d brukes til å konfigurere automatisk backup av virtuelle maskiner til en Recovery Services vault.",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-infrastructure/references/bcdr/capacity-planning-dr-configurations.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 17,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-infrastructure/bcdr/capacity-planning-dr-configurations.md#8",
"judge_verdict": "not_grounded",
"rule": "R3",
"evidence_url": "https://learn.microsoft.com/azure/cosmos-db/concepts-limits",
"evidence_quote": "| Minimum RU/s the system can scale to | `0.1 * Tmax` |",
"reason": "Autoscale har ingen konfigurerbar «baseline»-parameter — kun --max-throughput; med maks 10000 er gulvet 1000 RU/s, ikke 4000, så claimets ramme og tall stemmer ikke med gjeldende modell.",
"file": "skills/ms-ai-infrastructure/references/bcdr/capacity-planning-dr-configurations.md",
"line": 154,
"claim": "Cosmos DB-autoscale i DR-regionen konfigureres med baseline 4000 RU/s og maks 10000 RU/s (--max-throughput 10000).",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/capacity-planning-dr-configurations.md#9",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/azure/well-architected/design-guides/disaster-recovery",
"evidence_quote": "These systems generally target SLOs around 99.9%, with RTO and RPO measured in hours. An active-passive with warm/cold deployment strategy is common",
"reason": "Cold standby oppgis i kilden til timer (Foundry-siden sier «Hot/cold: 2 to 8 hours»), ikke 30–60 min; RTO-tallene er selve påstanden her, så den bærende delen er motsagt.",
"file": "skills/ms-ai-infrastructure/references/bcdr/capacity-planning-dr-configurations.md",
"line": 176,
"claim": "DR-strategier med tilhørende RTO: Active-Active (full) ~0 | Active-Active (optimized autoscale) sekunder | Warm Standby (partial) 5–15 min | Cold Standby (IaC only) 30–60 min | Backup & Restore timer–dager.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/capacity-planning-dr-configurations.md#12",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/azure/search/search-region-capacity",
"evidence_quote": "AZ support is relevant when the workload requires a 99.99% query SLA or operates in a regulated industry with documented high-availability requirements.",
"reason": "Replika-delen (2 = read-only SLA, 3 = read-write SLA) stemmer, men den bærende påstanden om at det ikke finnes noen 99,99 %-tier motsies av at Learn omtaler et 99,99 % query-SLA.",
"file": "skills/ms-ai-infrastructure/references/bcdr/capacity-planning-dr-configurations.md",
"line": 193,
"claim": "For Azure AI Search gir 2 replikaer kun read-only (query) SLA, mens 3 replikaer gir full read-write (query + indeksering) SLA; begge SLA-nivåene er 99,9 % og det finnes ingen 99,99 %-tier.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/capacity-planning-dr-configurations.md#13",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/azure/cosmos-db/autoscale-faq",
"evidence_quote": "Does autoscale work with reserved capacity? Yes. With reserved capacity for accounts with single-write regions, the reservation discount for autoscale resources is applied to the meter usage at a ratio of 1.5 times",
"reason": "Raden «Cosmos DB (autoscale) = ingen RI (per-bruk)» er direkte motsagt, og AI Search står ikke i Azure Reservations sin oversikt over dekkede tjenester; den siterte WAF-siden nevner kun «reserved instances or Azure Hybrid Benefit» generelt.",
"file": "skills/ms-ai-infrastructure/references/bcdr/capacity-planning-dr-configurations.md",
"line": 313,
"claim": "Reservasjonsanbefalinger for DR-ressurser: App Service P2v3 = 1-år RI for baseline | Cosmos DB (autoscale) = ingen RI (per-bruk) | Azure OpenAI PTU = RI kun for primær | AI Search Standard = RI for begge regioner | Storage (GZRS) = reservert kapasitet.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/capacity-planning-dr-configurations.md#14",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/en-us/azure/cost-management-billing/reservations/save-compute-costs-reservations",
"evidence_quote": "Reserved Virtual Machine Instance ... Azure Blob storage reserved capacity ... Azure Files reserved capacity ... Azure Cosmos DB reserved capacity ... Azure Synapse Analytics ... App Service stamp fee ... Azure NetApp Files",
"reason": "At partisjoner alltid kjører er greit, men den kanoniske opplistingen av hva Azure Reservations dekker inneholder ikke Azure AI Search — anbefalingen om reservasjon i begge regioner forutsetter et tilbud som ikke er dokumentert.",
"file": "skills/ms-ai-infrastructure/references/bcdr/capacity-planning-dr-configurations.md",
"line": 316,
"claim": "AI Search-partisjoner kjører alltid og bør derfor dekkes av reservasjon i begge regioner.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/capacity-planning-dr-configurations.md#17",
"judge_verdict": "not_grounded",
"rule": "",
"evidence_url": "https://learn.microsoft.com/en-us/azure/search/search-capacity-planning",
"evidence_quote": "You can add or remove replicas and partitions independently to accommodate fluctuating workloads.",
"reason": "Kilden sier at partisjoner kan økes og reduseres når som helst («Use the slider to increase or decrease the number of partitions»), altså det motsatte av at nedskalering krever gjenoppretting av tjenesten.",
"file": "skills/ms-ai-infrastructure/references/bcdr/capacity-planning-dr-configurations.md",
"line": 354,
"claim": "AI Search-partisjoner kan ikke skaleres ned uten å gjenopprette tjenesten, derfor må partisjoner dimensjoneres identisk i begge regioner.",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-infrastructure/references/bcdr/chaos-engineering-ai-systems.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 25,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-infrastructure/bcdr/chaos-engineering-ai-systems.md#6",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/azure/chaos-studio/chaos-studio-region-availability",
"evidence_quote": "| Region | Chaos Studio (Experiments) | Resource Targeting| | East US | Available | Available | ... | West Europe | Available | Available | | North Europe | Not Available | Available | | Sweden Central | Available | Available | | Germany West Central | Not Available | Available | | France Central | Not Available | Available | | Italy North | Not Available | Available | | Brazil South | Available | Available | | Australia East | Available | Available |",
"reason": "Jeg hentet den kanoniske regionslisten som ville ha oppført regionen, og Norway East (norwayeast) er ikke til stede i den i det hele tatt — verken for eksperimenter eller ressursmålretting — så påstanden om at eksperimenter kan opprettes i norwayeast er motbevist av fraværet.",
"file": "skills/ms-ai-infrastructure/references/bcdr/chaos-engineering-ai-systems.md",
"line": 66,
"claim": "Chaos Studio-eksperimenter kan opprettes i regionen norwayeast (Norway East).",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/chaos-engineering-ai-systems.md#20",
"judge_verdict": "source_silent",
"rule": "",
"evidence_url": "https://learn.microsoft.com/azure/chaos-studio/chaos-studio-overview",
"evidence_quote": "",
"reason": "Ingen learn.microsoft.com-side setter opp denne verktøylisten; Litmus Chaos og Toxiproxy er tredjepartsverktøy Microsoft Learn ikke er forventet å katalogisere, så listen kan verken bekreftes eller motbevises.",
"file": "skills/ms-ai-infrastructure/references/bcdr/chaos-engineering-ai-systems.md",
"line": 372,
"claim": "Verktøy for chaos testing mot Azure: Azure Chaos Studio | Azure Load Testing | Litmus Chaos | Toxiproxy | PYRIT.",
"disposition": "unsourced"
}
]
},
{
"file": "skills/ms-ai-infrastructure/references/bcdr/compliance-requirements-bcdr.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 10,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-infrastructure/bcdr/compliance-requirements-bcdr.md#4",
"judge_verdict": "not_grounded",
"rule": "R2",
"evidence_url": "https://learn.microsoft.com/azure/reliability/regions-paired",
"evidence_quote": "| Norway East | Norway West |",
"reason": "Den kanoniske region-par-listen gir Norway West som Norway East sin sekundær-/DR-region (og Sweden Central er paret med Sweden South); Sweden Central står ikke oppført som sekundærregion for Norway East noe sted på Learn.",
"file": "skills/ms-ai-infrastructure/references/bcdr/compliance-requirements-bcdr.md",
"line": 93,
"claim": "Sweden Central er typisk DR-region (sekundærregion) for Norway East.",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-infrastructure/references/bcdr/cost-analysis-dr-configurations.md",
"batch": "R7.4",
"judged_at": "2026-07-31",
"per_file_verdict": "flagged",
"claim_count": 16,
"verified": null,
"verified_by": null,
"flags": [
{
"id": "ms-ai-infrastructure/bcdr/cost-analysis-dr-configurations.md#2",
"judge_verdict": "not_grounded",
"rule": "R4",
"evidence_url": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule",
"evidence_quote": "| gpt-4o | 2024-05-13 | Deprecated | 2026-10-01 | gpt-5.1 |",
"reason": "Den gjeldende raden viser at alle gpt-4o-versjoner (2024-05-13, 2024-08-06, 2024-11-20) står som Deprecated - livssyklussiden definerer det som ikke lenger tilgjengelig for nye kunder - med utfasing fra 2026-10-01, så påstanden om at GPT-4o er tilgjengelig er superseded.",
"file": "skills/ms-ai-infrastructure/references/bcdr/cost-analysis-dr-configurations.md",
"line": 120,
"claim": "GPT-4o er tilgjengelig som modell i Azure OpenAI og brukes som primærmodell i TCO-eksempelet.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/cost-analysis-dr-configurations.md#13",
"judge_verdict": "not_grounded",
"rule": "R4",
"evidence_url": "https://learn.microsoft.com/en-us/azure/cdn/classic-cdn-retirement-faq",
"evidence_quote": "Starting August 15, 2025, Azure CDN from Microsoft (classic) no longer supports new resource creation, new domain onboarding, or managed certificates.",
"reason": "Gjeldende status er at Azure CDN Standard from Microsoft (classic) utfases 30. september 2027 og ikke lenger kan opprettes; Microsoft peker på Azure Front Door Standard/Premium som CDN-plattformen, så påstanden om at Azure CDN er tilgjengelig som CDN-tjeneste er superseded.",
"file": "skills/ms-ai-infrastructure/references/bcdr/cost-analysis-dr-configurations.md",
"line": 269,
"claim": "Azure CDN er tilgjengelig som CDN-tjeneste for å cache statisk innhold (dokumenter) og redusere cross-region-trafikk.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/cost-analysis-dr-configurations.md#14",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/cli/azure/costmanagement/export",
"evidence_quote": "--storage-account-id The ID of the storage account to store exports.",
"reason": "Parameteren heter --storage-account-id, ikke --storage-account; øvrige deler (--name, --scope, --type ActualCost, --timeframe MonthToDate, --storage-container, --recurrence Monthly, --recurrence-period) stemmer, men den ene bærende strengen er feil og ville feilet ved kopiering.",
"file": "skills/ms-ai-infrastructure/references/bcdr/cost-analysis-dr-configurations.md",
"line": 279,
"claim": "Azure CLI-kommandoen az costmanagement export create støtter parametrene --name, --scope, --type med verdien ActualCost, --timeframe med verdien MonthToDate, --storage-account, --storage-container, --recurrence med verdien Monthly og --recurrence-period.",
"disposition": "outdated"
},
{
"id": "ms-ai-infrastructure/bcdr/cost-analysis-dr-configurations.md#15",
"judge_verdict": "not_grounded",
"rule": "R8",
"evidence_url": "https://learn.microsoft.com/en-us/cli/azure/consumption/budget",
"evidence_quote": "az consumption budget create --amount --budget-name --category {cost, usage} --end-date --start-date --time-grain {annually, monthly, quarterly} [--acquire-policy-token] [--change-reference] [--meter-filter] [--resource-filter] [--resource-group] [--resource-group-filter]",
"reason": "az consumption budget create har verken --time-period, --notifications eller --resource-groups (de hører til create-with-rg/update), og bruker --start-date/--end-date samt små bokstaver i verdiene cost og annually, så flere bærende parameternavn og verdier er feil.",
"file": "skills/ms-ai-infrastructure/references/bcdr/cost-analysis-dr-configurations.md",
"line": 290,
"claim": "Azure CLI-kommandoen az consumption budget create støtter parametrene --budget-name, --amount, --category med verdien Cost, --time-grain med verdien Annually, --time-period, --resource-groups og --notifications med operator GreaterThan og threshold-verdier.",
"disposition": "outdated"
}
]
},
{
"file": "skills/ms-ai-security/references/ai-security-engineering/ai-prompt-shield-network.md",
"batch": "R7.1",