Systematicly is built on the principle that your research data should never be used to train AI models, never be accessible to other users, and always be under your control.
Every layer of Systematicly is designed with data security as a first principle, not an afterthought.
Systematicly uses the Anthropic commercial API, which is contractually prohibited from using your inputs or outputs to train generative models. Your research questions, uploaded PDFs, extracted data, and generated reports are never fed into any training pipeline. This is not an opt-out. It is the default and only behaviour for commercial API access.
All data is encrypted using AES-256 at rest and TLS 1.3 in transit. Your files, session data, and credentials are protected by industry-standard encryption at every stage. Data flows from your browser to our servers to the database.
Under Anthropic's commercial API terms, inputs and outputs are retained for a maximum of 30 days solely for trust and safety monitoring, not for training or product improvement. After processing, your research data exists only in your own account storage. You can delete your data at any time.
Each user's data is logically isolated. Your projects, uploads, and analysis results are never accessible to other users. Authentication is handled with row-level security policies that enforce strict data boundaries at the database level.
Every Systematicly account requires two-factor authentication (2FA) at sign-up. This adds a second layer of verification beyond your password, ensuring that even if credentials are compromised, your research data remains protected. We support authenticator apps and email-based verification codes.
You can export all your data at any time as a JSON archive (projects, settings, and analysis history). You can also request complete deletion of your account and associated data. We collect only the minimum data necessary to operate the platform.
A transparent look at how data flows through Systematicly.
PDFs, research questions, and search terms are encrypted with TLS 1.3 during transit.
Data is encrypted at rest (AES-256) with row-level security. Only your authenticated session can access it.
Only when you click "Analyse", "Extract", "Generate", etc. The minimum necessary data is sent via encrypted connection to Anthropic's commercial API.
Anthropic processes your request and returns results. Under commercial API terms, your data is not used for training, retained for max 30 days for safety monitoring only, then permanently deleted.
Your data is never used to train or improve AI models. It's never shared with third parties. It's never accessible to other users. It's never sold or monetised.
For organisations handling clinical data, commercially sensitive research, or requiring compliance with specific regulatory frameworks, we offer enterprise security arrangements including zero data retention.