Marburg, Germany

Philipps-Universität Marburg

Last checkedMay 20, 2026Last changedMay 20, 2026Review: Agent reviewedConfidence92%

Public JSON Change log Monthly report Citation rules Dataset access Analysis JSON Submit correction

Policy statusReviewed evidence-backed recordReview: Agent reviewedClaim coverage4 reviewedEvidence-backed claims4Reviewed4Candidate0Official sources3Source languagedePublic JSON/api/public/v1/universities/philipps-universitat-marburg.json

Coverage score90/100Coverage labelbroad public coverageReview: Machine candidateAnalysis confidence76%

Policy presence

Philipps-Universität Marburg has 2 source-backed public claims for policy presence; deterministic analysis status: unclear.

UnclearMachine candidateConfidence75%Evidence2Sources2

Claim claim-umr-teaching-generative-ai-not-prohibited-20260520 Claim claim-umr-fb03-ai-disclosure-deception-20260520

AI disclosure

Philipps-Universität Marburg has 1 source-backed public claim for ai disclosure; deterministic analysis status: recommended.

RecommendedMachine candidateConfidence78%Evidence1Sources1

Claim claim-umr-hrz-ki-service-rules-20260520

Coursework

Philipps-Universität Marburg has 3 source-backed public claims for coursework; deterministic analysis status: blocked.

BlockedMachine candidateConfidence75%Evidence3Sources2

Claim claim-umr-teaching-generative-ai-not-prohibited-20260520 Claim claim-umr-ai-generated-text-marking-20260520 Claim claim-umr-fb03-ai-disclosure-deception-20260520

Exams

Philipps-Universität Marburg has 3 source-backed public claims for exams; deterministic analysis status: blocked.

BlockedMachine candidateConfidence75%Evidence3Sources2

Claim claim-umr-teaching-generative-ai-not-prohibited-20260520 Claim claim-umr-ai-generated-text-marking-20260520 Claim claim-umr-fb03-ai-disclosure-deception-20260520

Privacy and data entry

Philipps-Universität Marburg has 2 source-backed public claims for privacy and data entry; deterministic analysis status: blocked.

BlockedMachine candidateConfidence77%Evidence2Sources2

Claim claim-umr-hrz-ki-service-rules-20260520 Claim claim-umr-teaching-generative-ai-not-prohibited-20260520

Academic integrity

Philipps-Universität Marburg has 2 source-backed public claims for academic integrity; deterministic analysis status: required.

RequiredMachine candidateConfidence74%Evidence2Sources2

Claim claim-umr-ai-generated-text-marking-20260520 Claim claim-umr-fb03-ai-disclosure-deception-20260520

Approved tools

No source-backed public claim identifying approved or licensed AI tools is present in this profile.

The current public tracker record does not contain claim evidence that identifies institutionally approved, licensed, procured, or enterprise AI tools.

Not MentionedMachine candidateConfidence0%Evidence0Sources0

Named AI services

Philipps-Universität Marburg has 1 source-backed public claim for named ai services; deterministic analysis status: restricted.

RestrictedMachine candidateConfidence78%Evidence1Sources1

Claim claim-umr-hrz-ki-service-rules-20260520

Teaching guidance

Philipps-Universität Marburg has 2 source-backed public claims for teaching guidance; deterministic analysis status: recommended.

RecommendedMachine candidateConfidence76%Evidence2Sources1

Claim claim-umr-teaching-generative-ai-not-prohibited-20260520 Claim claim-umr-ai-generated-text-marking-20260520

Research guidance

No source-backed public claim about research AI use is present in this profile.

The current public tracker record does not contain claim evidence about research use, publication ethics, research data, grants, or human-subjects compliance.

Not MentionedMachine candidateConfidence0%Evidence0Sources0

Security and procurement

No source-backed public claim about AI security review or procurement is present in this profile.

The current public tracker record does not contain claim evidence about security review, procurement, vendor approval, risk assessment, authentication, SSO, or enterprise licensing.

Not MentionedMachine candidateConfidence0%Evidence0Sources0

Derived tool records0

No tool-level evidence is published for this record yet. Broad AI tool mentions are not expanded into named tool conclusions.

Evidence-backed claims

4 reviewed evidence-backed public claim

Normalized value: gwdg_ki_cloud_rules_no_sensitive_data_transparency

Last checked: May 20, 2026
Last changed: May 20, 2026

Original evidence

Evidence 1

Geben Sie niemals personenbezogene, vertrauliche oder sicherheitsrelevante Daten über den Prompt ein bzw. stellen sie diese niemals dem genKI-Modell über Dateien zur Verfügung. ... Sollte der Dienst genutzt werden, muss dies transparent angegeben werden.
Source: KI-gestützte Dienste (GWDG KI-Cloud)

Localized display only

Do not enter personal, confidential, or security-relevant data through prompts or files; if the service is used, this must be transparently stated.

Source language: de
Snapshot hash: d9589821d83cd949b8bb4f06069281d8bc475a3f009bcc372f0511eeb6c1c36c
Retrieved: May 20, 2026
Evidence locator: Wichtige Regeln zur Nutzung

Normalized value: central_guidance_not_general_ban

Last checked: May 20, 2026
Last changed: May 20, 2026

Original evidence

Evidence 1

Die UMR verbietet nicht den Einsatz generativer KI. Dazu gehören sprachbasierte KI-Anwendungen (z.B. ChatGPT) ebenso wie bildgenerierende KI-Anwendungen (z.B. DALL-E, Midjourney).
Source: Positionspapier Lehre und KI

Localized display only

UMR does not prohibit the use of generative AI, including language-based applications such as ChatGPT and image-generating applications.

Source language: de
Snapshot hash: 000c1200bd247b5d605806da233797aabd32d15af760bb867bad210875fdafa5
Retrieved: May 20, 2026
Evidence locator: TEIL 1: Positionspapier

Normalized value: ai_generated_text_parts_should_be_marked

Last checked: May 20, 2026
Last changed: May 20, 2026

Original evidence

Evidence 1

Gute wissenschaftliche Praxis erfordert, dass KI-generierte Textteile als solche gekennzeichnet werden.
Source: Positionspapier Lehre und KI

Localized display only

Good scientific practice requires that AI-generated text parts be identified as such.

Source language: de
Snapshot hash: 000c1200bd247b5d605806da233797aabd32d15af760bb867bad210875fdafa5
Retrieved: May 20, 2026
Evidence locator: Überlegungen zu schriftlichen Prüfungsleistungen

Normalized value: fb03_ai_derived_information_must_be_marked

Last checked: May 20, 2026
Last changed: May 20, 2026

Original evidence

Evidence 1

In jedem Fall müssen Informationen, die direkt oder sinngemäß aus einem solchen Tool stammen, gekennzeichnet werden, andernfalls handelt es sich hierbei um einen Täuschungsversuch.
Source: Umgang mit künstlicher Intelligenz im Rahmen von wissenschaftlichen Arbeiten

Localized display only

Information taken directly or indirectly from such a tool must be marked; otherwise this is a deception attempt.

Source language: de
Snapshot hash: 32fb0ea6b92d04bf1de633f68a54175cc2284405112ac759d17f644530600a26
Retrieved: May 20, 2026
Evidence locator: Generell gilt

Candidate claims

0 machine or needs-review claim

Official sources

3 source attribution

KI-gestützte Dienste (GWDG KI-Cloud)

uni-marburg.de

Source URL: https://www.uni-marburg.de/de/hrz/dienste/ki-dienste
Snapshot hash: d9589821d83cd949b8bb4f06069281d8bc475a3f009bcc372f0511eeb6c1c36c

Positionspapier Lehre und KI

uni-marburg.de

Source URL: https://www.uni-marburg.de/de/universitaet/lehre/lehre-und-ki
Snapshot hash: 000c1200bd247b5d605806da233797aabd32d15af760bb867bad210875fdafa5

Umgang mit künstlicher Intelligenz im Rahmen von wissenschaftlichen Arbeiten

uni-marburg.de

Source URL: https://www.uni-marburg.de/de/fb03/ekw/studium/umgang-mit-kuenstlicher-intelligenz
Snapshot hash: 32fb0ea6b92d04bf1de633f68a54175cc2284405112ac759d17f644530600a26

Last checkedMay 20, 2026Last changedMay 20, 2026Open change log

Open contribution paths Review workflow

Back to universities

Philipps-Universität Marburg

Record status

Policy profile

Policy presence

AI disclosure

Coursework

Exams

Privacy and data entry

Academic integrity

Approved tools

Named AI services

Teaching guidance

Research guidance

Security and procurement

AI tools

Evidence-backed claims

The HRZ AI services page says the university offers generative AI services via GWDG KI-Cloud and lists usage rules including no prompt/file input of personal, confidential, or security-relevant data and transparent disclosure when the service is used.