Security and privacy at Campus
How we protect your data, where it lives, and how we work with AI.
- Never used to train AI
- EU-hosted in Frankfurt
- GDPR-ready
- Encrypted
- Invite-only
The basics
Never used to train AI
Content you send to our AI coach is processed under terms that forbid using it to train AI models. Your learning stays your learning.
Hosted in the EU
All platform data and storage run in Frankfurt, Germany (AWS eu-central-1), in a single region. Apart from the AI-coach flow, nothing leaves the EU, and that flow is safeguarded and disclosed.
GDPR-ready
We maintain a published privacy policy and a signed Data Processing Agreement, and we support data-subject requests.
Encrypted
Encrypted in transit over TLS 1.2+ (HTTPS) and at rest with AES-256. Secrets and credentials stay server-side and are never exposed to the browser.
A closed platform
There is no open signup. Every account is created by invitation, so we always know exactly who is on the platform.
Access is scoped
Learners cannot reach one another's data. Access is role-based with least privilege, enforced at both the application and the data layer.
How we keep Campus secure
Security is something we test continuously, not a box we check once.
Continuous security testing
We run automated security checks every day, probing the platform for weaknesses so we can fix them before they reach production.
Attack prevention by design
Our AI coach is a single, controlled assistant rather than an open, user-steerable agent, which keeps the surface for prompt injection and data exfiltration small. The AI can only ever act on content a learner already has access to.
Dependencies and monitoring
We monitor our dependencies for known vulnerabilities and patch them promptly. Errors are reported to our team directly, with personal data suppressed.
Audit and activity logging
User actions are recorded, so changes are traceable and we can see who did what. This is operational and security logging, not a tool for monitoring individual learners.
Modern authentication
Sign-in is passwordless by default, using secure email links and one-time codes. Two-factor authentication is available, and role-scoped permissions give precise control over who can do what.
A modern, maintained stack
Campus is built on a modern, actively maintained technology stack. All data changes are validated server-side, so the integrity of a change never depends on the browser.
Privacy and your data
Who can see your data, what reaches the AI, and the control you keep over it.
Your conversations are private
Coaching conversations with Henrik are visible only to the learner. Trainers, leaders, and administrators see learning progress, never the content of a conversation.
Least-privilege access
Access is role-based and enforced at both the application and the data layer, so people only see what their role allows. Learners cannot reach one another's data, and administrative access to the underlying data is limited to our own team.
The AI receives only what it needs
You decide what to share with the coach; a sector and a general description of your team are enough. Content is processed under data-processing terms and is never used to train AI models.
Your data stays in your control
We process only what the platform needs to run. You can request export or deletion of personal data, and when a contract ends, data is deleted on a defined schedule.
More and advanced
For reviewers, procurement, and works councils.
Subprocessors
Legal and assurance
Our privacy policy is published in English and German. A signed Data Processing Agreement, the subprocessor list, and our responses to standard security questionnaires are available, and we work with an external Data Protection Officer. For German works councils, we are glad to walk through logging, retention, and confidentiality.
Talk to us
Campus developers Zakir, Ángel and Tom, and Philip (Head of Business Development), are available any time to answer your questions.
Frequently asked questions
- Is our data used to train AI?
- No. Content processed by the AI coach is handled under enterprise terms that prohibit using it for model training. Nothing on Campus feeds an AI training set.
- Where is our data stored?
- On AWS in Frankfurt, Germany, in a single EU region. The one exception is content you send to the AI coach, processed by Anthropic in the US under Standard Contractual Clauses, never used for training.
- Can my manager see my coaching conversations?
- No. Coaching conversations are visible only to the learner. Trainers, leaders, and administrators see learning progress, not conversation content.
- Can we use Campus without exposing who we are?
- Largely, yes. You can describe your context without identifying detail: give a sector instead of a company, and describe teams in general terms. Customers remain responsible for what their learners choose to enter.
- Do you have a Data Processing Agreement?
- Yes. We maintain a signed Data Processing Agreement alongside a published privacy policy and a subprocessor list.
- What about certifications?
- We build on infrastructure certified to ISO 27001, SOC 1/2/3, and C5 (AWS). These are our infrastructure provider's certifications; we do not currently hold or claim our own ISO 27001 or SOC 2 certification.