Trust & security
Straight answers about how your data is handled
Here's what's actually in place today, and — just as plainly — what isn't yet. We'd rather under-promise than imply a certification we don't hold. That goes double for AI: it's the one part of research where governance can't be an afterthought.
AI governance
AI on your terms
Surveti is built to be the research platform with enterprise-grade AI governance: every AI feature has an off switch, every AI decision is logged, and every run is metered on one transparent ledger. Governance isn't bolted on — it's how the AI works.
Full off switch
Every AI feature — survey import, adaptive probing, AI summaries — can be turned off entirely, with an organization-wide default and a per-workspace override. Teams under strict data policies can disable AI completely.
Your data never trains models
Surveti never uses your survey data to train AI models.
AI activity log
Every AI decision — which feature ran, when, which model, and token usage — is recorded, and workspace admins can view and export the full trail.
Metered and transparent
AI runs on one transparent credit ledger. No hidden AI usage — you can see exactly what ran.
Model transparency
AI features run on leading large language models, selected per feature. The models in use are documented.
In place today
- Encryption in transit with TLS on every connection
- Encryption at rest with AES-256; PII encrypted in the database
- Immutable audit logs of access and key actions
- GDPR and CCPA data deletion support, for account and respondent data
- Hosted on Amazon Web Services (AWS)
- AI-based fraud traps and duplicate/attention detection in the survey runtime
- Single sign-on (SSO) — available on Enterprise
- Role-based access with client viewer (read-only) seats
- A full off switch for AI — organization-wide default plus per-workspace override
- AI activity log: every AI action recorded, viewable and exportable by admins
- Your survey data is never used to train AI models
On the roadmap — not yet in place
We list these explicitly so nothing above is mistaken for a current guarantee. If one of these is a requirement for you, contact us and we'll share our honest timeline.
- SOC 2 Type II attestation
- Advanced organization admin and audit controls beyond current logs
- Contracted uptime SLAs outside of Enterprise agreements
- Additional regional data-residency options
Data ownership & deletion
You own the surveys and response data you create. Surveti processes it to run the service on your behalf. You can export your data at any time, and we support GDPR and CCPA deletion requests for both account data and survey response data. Details are in our Privacy Policy.
Need to complete a security questionnaire or vendor review? We'll answer it directly — email security@surveti.com.
AI governance, answered
Can we turn AI off?
Yes — entirely. AI features (survey import, adaptive probing, AI summaries) can be turned off with an organization-wide default and a per-workspace override, so teams under strict data policies can disable AI completely.
Do you use our data to train AI?
No. Surveti never uses your survey data to train AI models.
Can we see what the AI did?
Yes. Every AI decision — which feature ran, when, which model, and token usage — is recorded in an AI activity log that workspace admins can view and export.
How is AI usage billed?
AI runs on one transparent credit ledger. There is no hidden AI usage — you can see exactly what ran.
Which AI models does Surveti use?
AI features run on leading large language models, selected per feature. The models in use are documented.