AI for medical & scientific societies

Designing how people,
data and AI work together.

Artificial intelligence is creating new possibilities for medical and scientific societies — not by replacing the people who lead and manage them, but by reducing repetitive work, making organisational knowledge easier to use and connecting information that is often scattered across different systems.

A connected data environment

Useful information should not stay trapped in separate systems.

Membership, congresses, courses, journals, governance and finance all generate valuable information. Today, these data often sit in separate systems and are difficult to use across the organisation.

Twister is exploring a model in which these sources can be connected securely and selectively — without giving up data ownership or creating one uncontrolled central database.

Society Data Hub · SDH

Connect data.
Don’t centralise risk.

The Society Data Hub is the controlled layer that connects relevant information across the organisation. It does not necessarily move all data into one database. Instead, it can provide governed access to existing systems, based on defined permissions, responsibilities and purpose.

Members

Member Data Hub (MDH)

Authoritative membership data, roles, status, contact information and permissions.

Events & education

Congresses & Courses

Registrations, participation, faculty, programmes, fees, CME and evaluation data.

Publishing

Journal Data

Society-accessible publication, editorial and performance data, subject to publisher agreements and data rights.

Institutional knowledge

Governance & Knowledge

Minutes, policies, contracts, guidelines, board decisions and organisational history.

Management

Finance & Sponsorship

Budgets, payments, commitments, sponsorships and management reporting where access is authorised.

Controlled connection layer

Society Data Hub SDH

One governed access model across the society’s information landscape.

PermissionsAudit trailData ownershipRetention rulesPurpose limitationSecure APIs
Executive OfficeBriefings · follow-ups · coordination
Member ServicesRequests · renewals · communication
Congress & EducationOperations · reporting · workflows
Knowledge & GovernancePolicies · minutes · decisions
Journal IntelligenceAuthorised publishing insights
Finance & SponsorshipReporting · commitments · oversight
Human governance layerStaff · Executive Committee · Board

AI supports work. People retain judgement, approval and accountability.

Data sovereignty & security

Your data remains your data.

AI should not require a society to give up control of its information. Security, privacy and governance need to be designed into the workflow — not added afterwards.

01

Society ownership

The society remains in control of its source data, access rules and retention decisions.

02

Minimum access

Each workflow or agent should receive only the data and permissions required for its defined task.

03

Source-system first

Where practical, data stay in the systems designed to hold them rather than being copied into an unnecessary central pool.

04

Human approval

Sensitive changes, external communication and high-impact actions can require explicit human review.

05

Traceability

Access, actions and important workflow steps should be logged so activity can be reviewed and understood.

06

Privacy & compliance

Architectures should reflect GDPR and applicable Swiss data-protection requirements, contracts and processor obligations.

A practical design principle

For an early deployment, Twister would favour read-only or tightly limited access wherever possible. Write access can be introduced later for defined workflows, with approval rules and clear auditability.

From architecture to useful work

Start with one problem that matters.

The opportunity is not one large AI project. It is a sequence of focused improvements to the way the society works — each with a defined purpose, measurable benefit and appropriate controls.

1

Map

Identify systems, information flows, data ownership and pain points.

2

Select

Choose a focused use case where AI can create practical value.

3

Protect

Define permissions, data handling, human review and escalation rules.

4

Test & scale

Measure the result, refine the workflow and expand only where it works.

AI-enabled society management

Where could your society remove friction without giving up control?

We can start with one recurring administrative problem, one information bottleneck or one workflow that has become unnecessarily complex.

Start a conversation →