Guides
Answers for everyone introducing AI in a hospital or public administration. With figures, sources and date.
No guide matches this selection.
Selection
- What is an AI platform? Platform, model and computing power
- SOC 2, ISO 27001, C5: Which proof counts for hospitals and public administration?
- Exit strategy for AI platforms: what to settle for leaving before you sign
- AI platform or enterprise search tool?
- Horizontal AI platform or AI in specialist systems: what hospitals need
- Wrapper or standalone platform?
- AI in laboratory diagnostics: how to assess specialist vendors such as medicalvalues
- AI platform or workflow tool such as n8n and Dify: an overview
- Tendering an AI platform: the guide in seven steps
- Procurement law for AI in Austria: what the BVergG 2018 requires since 2026
- Public procurement of AI in Switzerland: BöB for the Confederation, IVöB for cantons and municipalities
- Proof of concept for AI: a reliable decision in six weeks
- AI vendor landscape in Austria: who offers what for municipalities, healthcare and businesses?
- AI vendor landscape in France: who offers what for local authorities and healthcare?
- AI vendor landscape in Switzerland: who offers what for municipalities, hospitals and businesses?
- Anonymisation proxies and confidential AI: market overview for DACH and France
- Bus factor: when only one person knows the self-built AI
- Criteria catalogue for AI platforms: how to weight and evaluate providers
- From AI prototype to production system: what changes between test and operation
- Replacing the HIS after the end of SAP IS-H and i.s.h.med: what hospitals must decide now
- Is a Mac mini with Ollama enough for AI in your organisation?
- ISO 27001 and NIS2: what self-built AI means for your ISMS
- Microsoft Copilot or your own enterprise AI?
- Using open AI models in your organisation: what the licences allow
- Open WebUI or an AI platform: when is the interface enough?
- Scaling Ollama and Open WebUI: what changes when many people ask questions
- Security updates for a self-built AI stack: what is involved
- Sovereign AI for local authorities compared
- What open-source AI does not automatically include
- Why pay when there is a free government AI solution?
Costs
- From how many people does an AI server pay off compared with the cloud?
- Licensing models for AI platforms compared
- TCO comparison: run AI yourself or buy a platform?
- Funding for AI projects in Austria and Switzerland
- Calculating the ROI of AI projects: method, studies, worked example
- ratgeberKostenin progress
Law and data protection
- Is an AI platform a medical device?
- Traceable AI: audit log, sources and explainable AI
- AI literacy under Article 4 EU AI Act: train your staff yourself, with slides and templates
- Austria: ELGA and AI access
- EU AI Act for hospitals
- EU AI Act for organisations: the overview
- AI and the GDPR: what organisations need to know
- AI and the Swiss Federal Act on Data Protection (revFADP)
- When does an AI doctor's letter generator become a medical device?
- Transparency obligations under Article 50 of the EU AI Act
- Anonymisation or your own AI infrastructure: what protects data when using AI
- Austria: DSG, official secrecy and AI
- Austria: ID Austria and AI applications
- CE marking for AI diagnostic support
- Data processing agreements with AI providers: what to look for in the contract
- DPIA for AI: template and process
- EU AI Act for local authorities
- France: GDPR, CNIL and AI
- France: HDS certification for health data
- France: SecNumCloud
- Fundamental rights impact assessment (Art. 27 AI Act)
- Identifying high-risk AI: Annex III and Article 6
- Is pseudonymised use of AI automatically GDPR-compliant?
- Liability: self-built from open source or a purchased platform?
- Own AI apps and the medical device trap (Article 5(5) MDR)
- Do our own AI apps make us a provider under the AI Act?
- Professional secrecy in Switzerland (Art. 321 StGB) and AI
- Re-identification in pseudonymised AI requests: what studies show
- Record of processing activities for AI: what belongs in it
- Switzerland: Swissmedic and in-house medical devices
- Third-country transfers and AI: when a pseudonymisation proxy is not enough
- US CLOUD Act: what it means for AI in Europe
- Zero Data Retention: Is it enough for data protection?
Technology
- Coding with AI when cloud tools are not allowed
- Instance model and model hosting: two questions to keep apart
- Scaling an AI platform: larger models or more people?
- Hybrid AI: external models such as Claude for selected groups, sensitive data kept protected
- Mirth Connect is no longer open source: licence, forks and connecting AI
- Securing AI agents at runtime: what a valid sign-in does not cover
- Who sees what? Permissions for AI that searches documents
- AI in the ticketing system: what it can take on at the service desk
- Cloud or server on your premises: which fits which organisation?
- Confidential computing for AI: what TEE protection achieves and where its limits are
- AI and SAP: how an AI platform reaches the ERP system
- Context loss through anonymisation: why AI answers can get worse
- Hidden instructions in documents: How to protect AI from manipulation
- HIS, communication server, data platform, AI platform: who does what in the hospital?
- Hybrid operation: when does an AI platform on site with an external model pay off?
- Licence risk with open-source tools: three examples from 2025 and 2026
- Pseudonymisation proxy for ChatGPT and other AI services: how the technology works
- Taking roles and permissions for AI from your identity management
- AI without the internet: how does offline and air-gapped operation work?
- SPHN and BioMedIT: an overview of the Swiss research data infrastructure
- New AI model: how to check after a change that everything still works
- What is MCP? The Model Context Protocol explained for IT professionals
Introduction
- How does an AI model get its knowledge?
- What does IT have to do when we introduce an AI platform?
- AI policy for staff: structure, wording and a template to adopt
- AI and monitoring of staff performance or behaviour: where the line is
- Introducing AI without an IT department: what you do yourself and what others take on
- Acceptance of AI: how staff come to trust the new technology
- Change management for introducing AI: roles, phases and measuring success
- Onboarding and support for an AI rollout: what you can expect from the provider
- Organising AI training: who learns what, in which format and when
- Codetermination on AI: the rights of works councils, staff councils and church employee representatives