Safety
Defense-in-depth design, prohibited classroom outputs, fail-closed behavior, and adversarial testing principles.
Explore Safety →Schools should not have to guess what an AI product does, stores, or promises.
This Trust Center brings Splash Spectrum’s current safety, privacy, security, accessibility, and product-status commitments into one place. It distinguishes public-site features from pilot concepts and future classroom infrastructure.
Start with the question your school needs answered.
Security
Planned encryption, access controls, tenant isolation, audit logging, retention, and incident-response practices.
Explore Security →Privacy
Data minimization, limited educational purpose, school controls, provider transparency, and no behavioral advertising model.
Explore Privacy →Accessibility
WCAG 2.2 AA target, keyboard access, assistive-technology semantics, readable presentation, and reduced-motion support.
Explore Accessibility →Product Status
A plain-language view of what is live on the public site, what is pilot-stage, and what remains planned.
View Product Status →Terms
Current status of education-platform terms and why legacy personalized-book terms should not govern future classroom AI services.
View Terms Status →Trust requires precise language.
Not “100% safe”Not “zero harmful content”Not “FERPA certified”Not “COPPA certified”Not hidden chain-of-thought accessNot every roadmap feature is live
The production classroom platform will need legal, security, privacy, accessibility, and safety review against the system that actually ships. The public site intentionally separates design goals from completed capabilities.
Have a procurement, privacy, safety, or security requirement?
Bring it to us during the pilot-design process so the requirement can influence the product before broad deployment.
Contact Splash